'310^
HARVARD UNIVERSITY
Library of the
Museum of
Comparative Zoology
The Great Basin Naturalist
VOLUME 37, 1977
Editok: Stephen L. Wo
Plhlished at BaiciiAM YoL'.Nc; U.\i\i:usitv, hv Bru;ham Yolnc; U.ntvehsity
TABLE OF CONTEXTS
Volume 37 Xuinbcr 1 - March 3L 1977
ireeding biologv of the Douhle-cresled Ooiniorant on Ltali Lake. Houald M. Mit- chell \:. 1
>attail decline at Faiinini^ton Bay Waterfowl Manau;enient Area. Christine .Mmton
Anderson 24
0*' 1^
Osteology and myology of the head and neck regions of Callimunis, Cophosaurus,
Holbwokia, and (7?no (Reptilia: Iguanidae). Douglas C. Cox and Wilmer W.
Tanner 35
Hydrological features of a California coastal fen. Don C. Erman, Kenneth B. Roby,
and Michael Eames 57
Introduced and exported American Scolytidae (Coleoptera). Stephen L. Wood 67
The sagebrvish vole flea, Megabothris ckintoni princei, in western Utah, with com- ments on the distribution of Megabothris in the Bonneville Basin. Harold J.
Egoscue 75
Interpopulational variation of blood proteins in pika (Ochotona princeps saxatilis).
John T. Bninson, Richard N. Seaman, and Donald J. Nash 77
Observations on the distribution of microorganisms in desert soil. A. T. Volmer, F.
Au, and S. A. Bamberg 81
Home range of the western jumping mouse, Zapus princeps, in the Colorado Rocky
Mountains. Nathaniel Stinson, Jr 87
Preliminary mammal survey of Fossil Butte National Monument, Wyoming. Tim W.
Clark 91
An undescribed species of Thelypodiopsis (Brassicaceae) from the Uinta Basin, Utah.
Stanley L. Welsh and N. Duane Atwood 95
A mountain cercocarpus population— revisited. E. S. Nixon 97
A new status for Penstenion cyananthus Hook. ssp. subglaher Pennell (Scrophula-
riaceae). Stephen L. Clark 100
The influence of predator control on two adjacent wintering deer herds. Dennis D.
Austin, Philip J. Urness, and Michael L. Wolfe 101
Plants of Arizona: a new species of Astragalus from the Kaibab Plateau. Stanley L.
Welsh and Kaye H. Thome 103
Bionomics of two aspen bark beetles, Trypophloeus populi and Procryphahis mucro-
nattis (Coleoptera: Scolytidae). Jerold L. Petty 105
Food of townsend ground squirrels on the Arid Land Ecology Reserve (Washington).
MarkK. Johnson 128
Number 2 - June .30, 1977
Host-induced morphological variations in the strigeoid trematode Posthodiplostomiim minimum (Trematoda: Diplostomatidae). II. Body measurements and tegument modifications. James R. Palmieri 129
The subfamily Anomiopsyllinae (Hystrichopsyllidae: Siphonaptera). I. A revision of
the genus AnomiopsyUus Baker. A. M. Barnes, V. J. Tipton, and J. A. Wildie 138
New synonymy and new species of American bark beetles (Coleoptera: Scolytidae),
Part iV. Stephen L. Wood 207
Supplementary note on Venezuelan bat flies (Diptera: Nycteriobiidae). Lindolpho R.
Guimaraes 221
The systematics of Crotaphytus wislizeni, the leopard lizards. Part III. The leopard lizards of the Great Basin and adjoining areas, with a description of a new sub- species from the Lahontan Basin. Wilmer W. Tanner and Benjamin H. Banta .... 225
Syzygioxylon chhindwarense, a new fossil wood from the Deccan Intertrappean series
of India. E. M. V. Nambudiri and William D. Tidwell 241
Birds of Gunnison and Cub islands, Great Salt Lake, Utah. F. L. Knopf and V. C.
Bachman 247
Varying prevalence of Taenia taeniaeformis strobilocerci in Microtus pennsylvanicus
of Montana. Richard H. McBee 252
An annotated list of the aquatic insects of southeastern Idaho, Part II: Trichoptera.
Robert L. Newell and G. Wayne Minshall 253
A new status for Penstemon cyananthus Hook, ssp. subglaher Pennell (Scrophula-
riaceae). Stephen L. Clark 258
Range extension of Trionyx spiniferus enioryi into Utah (Reptilia). Carlos Y. Roby and
Earl C. Loveless 259
The life cycle of Hydroperla crosbyi (Plecoptera: Perlodidae). Reed Y. Oberndorfer
and Kenneth W. Stewart 260
Spring waterfowl migration in the Uinta Basin of northeastern Utah. Mary E. Sang-
ster 274
Number 3 - September 30, 1977
Utah flora: Brassicaceae (Cruciferae). Stanley L. Welsh and James L. Reveal 279
Typhlodromtis mcgregori Chant (Acarina: Phytoseiidae) and its control of phyto- phagous mites in Utah and southern California apple orchards. B. A. Croft and C. D. Jorgensen 366
Host-induced morphological variations in the strigeoid trematode Posthodiplostomum minimum (Trematoda: Diplostomatidae). III. Organs of attachment. James R. Palmieri 375
New synonymy and new species of American bark beetles (Coleoptera: Scolytidae),
Part V. Stephen L. Wood 383
Biomass of fifty conifer forests and nutrient exports associated with their harvest. T.
Weaver and F. Forcella 395
The algal flora of the campus of Brigham Young University, Provo, Utah. Cheng
Mou-Sheng and Samuel R. Rushforth 402
Occurrence of ticks on mule deer in central Utah. Jordan C. Pederson 407
Number 4 - December 31, 1977
Population dynamics of the Ferruginous Hawk during a prey decline. Neil D. Wof-
finden and Joseph R. Murphy 411
A new species of Rhopalomyia (Diptera: Cecidomyiidae) from Artemisia ludoviciana
Nuttall (Compositae) in Utah. Raymond J. Gagne 426
Bionomics of a subterranean gall midge (Diptera: Cecidomyiidae) from Artemisia lu- doviciana. M. A. S. K. Ranasinghe 429
The alpine butterflies of Castle Peak, Nevada County, California. Arthur M. Shapiro . 443
Geographic variation in Contopus sordidulus and C. virens north of Mexico. M. Ralph
Browning 453
Distributional patterns of mammals in Utah. David M. Armstrong 457
A new fossil sponge from the Ordovician Garden City limestone of southeastern
Idaho. J. Keith Rigby and James K. Gilland 475
Host-induced morphological variations in the strigeoid trematode Posthodiplostomum minimum (Trematoda: Diplostomatidae). IV. Organs of reproduction (ovary and testes), vitelline gland, and egg. James R. Palmieri 481
A survey of breeding and migratory birds southwest of Farmington, New Mexico.
Duane A. ToUe 489
A new monotypic genus of chiggers and four new species of Quadraseta from Vene- zuela (Acari: Trombiculidae). M. L. Goff and J. M. Brennan 501
A new name for Galium scahriusculum (Rubiaceae) of Utah. Lauramay T. Dempster
and Friedrich Ehrendorfer 510
New synonymy and new species of American bark beetles (Coleoptera: Scolytidae),
Part VI. Stephen L. Wood 511
Nectar-sugar concentrations and flower visitors in the western Great Basin. Larry J.
Gut, Robert A. Schlising, and Carol E. Stopher 523
New grass distribution records for New Mexico and the United States. Stephan L.
Hatch 530
Stilt Sandpiper and Hudsonian Godwit in Nevada. C. S. Lawson 532
Index to Volume 37 533
HE GREAT BASIN NATURALIST
)lume 37 No. 1 March 31, 1977
Brigham Young University
MUS. COMP. ZOOU.
LIBRARY
JAN20^b/o
HARVARD UNIVERSITY
GREAT BASIN NATURALIST
Editor. Stephen L. Wood, Life Science Museum, Brigham Young University, Provo, Utah 84602.
Assistant Editor. Stanley L. Welsh, Life Science Museum, Brigham Young University, Provo, Utah 84602.
Editorial Board. Kimball T. Harper, Botany; Wilmer W. Tanner, Life Science Museum; William D. Tidwell, Botany; Clayton M. White, Zoology.
Ex Officio Editorial Board Members. A. Lester Allen, Dean, College of Biological and Agri- cultural Sciences; Ernest L. Olson, Director, Brigham Young University Press, Univer- sity Editor.
The Great Basin Naturalist was founded in 1939 by Vasco M. Tanner. It has been pub- lished from one to four times a year since then by Brigham Young University, Provo, Utah. In general, only previously unpublished manuscripts of less than 100 printed pages in length and pertaining to the biological natural history of western North America are ac- cepted. The Great Basin Naturalist Memoirs was established in 1976 for scholarly works in biological natural history longer than can be accommodated in the parent publication. The Memoirs appears irregularly and bears no geographical restriction in subject matter. Manu- scripts are subject to the approval of the editor.
Subscriptions. The annual subscription rate of the Great Basin Naturalist is $12 (outside the United States the rate is $13). The price for single numbers is $4 each. All back num- bers are in print and are available for sale. All matters pertaining to the purchase of sub- scriptions and back numbers should be directed to Brigham Young University Press, Mar- keting Department, 204 UPB, Provo, Utah 84602. The Great Basin Naturalist Memoirs may be purchased from the same office at the rate indicated on the inside of the back cover of either journal.
Scholarly Exchanges. Libraries or other organizations interested in obtaining either jour- nal through a continuing exchange of scholarly publications should contact the Brigham Young University Exchange Librarian, Harold B. Lee Library, Provo, Utah 84602.
Manuscripts. All manuscripts or other copy for either the Great Basin Naturalist or the Great Basin Naturalist Memoirs should be addressed to the editor as instructed on the back cover.
© 1977 Brigham Young University Press. All rights reserved
Printed in United States of America
9-77 650 24964
The Great Basin Naturalist
PuBLisHi:i) AT Pro\o, Utah, by Brigham Young University
Volume 37
March 31, 1977
No. 1
BREEDING BIOLOGY OF THE DOUBLE-CRESTED CORMORANT ON UTAH LAKE'
Ronald M. Mitchell-
Abstract.— Research on the nesting of the Donble-crested Cormorant (Phalacrocorax auritus auritus) was un- dertaken to determine the breeding biology of the bird on Utah Lake, Utah County, and the possible effects that diking Provo Bay would have on the cormorants breeding there. In 1973, two colonies in Provo Bay were visited weekly to determine laying dates, incubation period, clutch size, and hatching dates. Another colony, located on the dike of the Geneva Steel Works Reservoir near Oreni, was visited daily during the spring and summer of 1973. At this colony egg-laying occurred from 13 April to 17 May. The average clutch size was 3.8, and the egg length and width averaged 60.50 mm x 39.05 mm. The average period of incubation was 28 days, and 29.5 per- cent of the eggs hatched, beginning on 20 May and ending on 9 June. The young weighed an average of 36 gm at hatching with a beak length of 17 mm and tarsometatarsal length of 11 mm. These measurements increased to 1543 gm, 73 mm, and 58 mm, respectively, in 23 days. The cormorant population on Utah Lake may be severely reduced by the proposed diking of Provo Bay.
There have been no detailed studie.s of the biology of the Double-crested Cormo- rant {Phalacrocorax auritus auritus) in Utah, other than a brief but noteworthy account by Behle (1958) of those birds that nested on Egg Island, Great Salt Lake. Two de- tailed studies of this subspecies have been conducted on the Atlantic coast (Lewis 1929, Mendall 1936), but no extensive ac- counts of breeding biology exist for Double- crested Cormorants found nesting inland. This lack of scientific data may be due in part to a lack of qualified and interested ob- .servers, but I feel it is principally due to the sentiment expressed in the following quotation:"God no doubt had his reasons for creating each living thing, but when he cre- ated the cormorant he did himself little credit" (Demille 1926). This attitude may account for the fact that the cormorant is
likely the least understood water bird nest- ing in Utah.
Although there are three other subspecies (P. a. albociliatus, P. a. floridanus, P. a. cin- cinatus) found in North America, P. a. au- ritus is considered the most important due to its wide distribution, ranging from central Canada to Texas and from the Atlantic coast to Utah. The P. a. auritus populations in Utah are the westernmost breeding colo- nies found in the United States and provide excellent contrast for compari.son with the previously studied birds nesting along the Atlantic coast.
The few publi.shed references that exist concerning Utah's cormorants are little more than sight records or brief mentions of nesting sites. Behle (1958) summarized the history of the seven cormorant colonies of which he was aware. At the time of his
This manuscripl was prepared in partial fulfillment of the requirements for the degree of Master of Science in Zoolojjy at Brighi Provo, L'tah. The research was supported in part by a grant from the National Audubon .ScK-iety.
Young University.
PO. Box 28147, San Ant(
Great Basin Naturalist
Vol. 37, No. 1
writing, all but two (Egg Island and Bear River) had been deserted or destroyed (Fig. 1). In addition I found a reference to an un- reported colony located on White Rock, a small island located a few hundred yards northwest of Antelope Island, Great Salt Lake. A set of cormorant eggs was collected from there on 10 May 1901 (Woodbury et al. 1949). I also discovered an unreported colony located west of Bountiful in 1973.
Behle (1958) did not mention Utah Lake as a nesting area, and it is mentioned only once in the other literature as a "former" nesting site (Woodbury et al. 1949). How- ever, I was able to find unpublished refer- ences that included a record of four cormo- rant eggs collected in Provo Bay in 1920 (Bee 1920). Cottam (1927) stated that the Double-crested Cormorant nested on Utah Lake. In 1937 John Hutchings of Lehi (Bee 1937) reported that several sets of Double- crested Cormorant eggs were collected on Rock Island in Utah Lake (Fig. 2). James Bee (1946) found six active cormorant nests in a tree on the Geneva dike in May 1946 and stated that the birds had nested there for two or three years. Bee (1949) reported
|
V/Be« |
|||
|
r^ A Logan |
/ |
||
|
r ^ p^ Brigham |
|||
|
^ Great \ 1 Salt a\ Y*"^^ Bou„t..uX |
UTAH |
||
|
^-N,_y salt Lake |
|||
|
■Rjoele |
|||
|
9- |
vernal |
||
|
Nephi |
price |
|
Clear LaKe ij » Fi Cedar Ci St. Ge9rge |
ty |
Moab |
|||
|
A Deserte • Present |
<. or destroyed coloni |
||||
a trip to a large cormorant colony in Provo Bay, south of the Provo airport dike, that contained 150 pairs of nesting cormorants.
The purposes of this study were to obtain information on the breeding biology of the unreported colonies on Utah Lake, obtain data on the nesting ecology of inland nest- ing cormorants for comparison with the data already published concerning coastal nesting birds, determine the current status of Utah's cormorant population, and gather information to determine the impact on the nesting sites of the two colonies located in Provo Bay by the proposed plan of the Bu- reau of Reclamation to dike and drain the bay as part of the Central Utah Project.
The Study Area
of
There are three nesting colonies Double-crested Cormorants located near Utah Lake. Two of the colonies are located in Provo Bay (Fig. 2), a shallow arm on the eastern side of Utah Lake. These have been designated the Springville and Bay colonies.
■ Springville Colo ^ Bay Colony ^ Geneva colony
Fig. 1. Location of the deserted and active nesting Fig. 2. Location of the three study sites on Utah
colonies in Utah. Lake.
March 1977
Mitchell: Cormorant Biology
The water depth in this area varies from a few milhmeters to 2 m and averages about 1 m. The principle vegetative types are hardstem buHrush (Scirpus acutiis), tamarix (Tamarix petandra), cottonwood {Populus fremonti), and willow (Salix amygdaloides). The third colony is located 11 km north of Provo Bay at the United States Steel Com- pany Geneva Works and is designated Gen- eva Colony (Fig. 2).
The Springville Colony is located 0.5 km due west of the mouth of Hobble Creek in a stand of cottonwood trees that range from 10-20 m in height and extend in a straight line west from the shore into the bay. The tree line, which is about 0.5 km long, was established in the late 1930's during low lake levels. The westernmost trees are dead, in 1 m of water, and in 1971 and 1972 the trees most distant from shore were used by cormorants for nesting. In 1973, apparently due to increased numbers, the birds moved eastward into some living cottonwoods that had previously been used exclusively by Great Blue Herons {Ardea herodias) for nesting.
The Bay Colony is located at the mouth of Provo Bay (Fig. 2), 2 km due south of the Provo airport. This site consists of a stand of cottonwoods and willows growing in an east-west line in about 0.6 m of wa- ter. In 1971 the area consisted of 18 trees ranging from 10-20 m in height. In 1973 cormorants nested with Great Blue Herons in one cottonwood and in four willow trees at the east end of the tree line.
The Geneva Colony is located on a diver- sion dike, built in 1940, that extends about 1.2 km within the reservoir. This artificial pond is located west of the steel plant proper and is about 0.2 km from Utah Lake. The water is contained by 3 m high earthen walls (Fig. 3). The water, which is over 6 m deep in most places, is used for cooling purposes in the steel-making pro- cess. The cormorants nested in one cotton- wood and three Siberian elms {Ulmiis pii- mila), all living, at the distal end of the dike. Great Blue Herons, Snowy Egrets {Leucophoyx thula), and Black-crowned Night Herons (Nycticorax nycticorax) nested in the same trees, while thousands of Cali-
fornia Gulls {Larus californicus) nested on the dike itself.
Methods and Procedures
The study began in the spring of 1972 and continued until the spring of 1974.
Two blinds were set up at the Geneva Colony to allow observation of nesting ac- tivity. Both blinds were set up on the east bank of the Geneva pond. One blind was placed opposite trees 1 and 2 about 12 m distant (Fig. 3). Six cormorant nests were visible from this blind. Tree 3, which con- tained three cormorant nests, was also ob- servable when binoculars were used. The second blind was placed about 300 m far- ther north near the northeast corner of the pond. Tree 4 was about 30 m away, and nine cormorant nests were observable when binoculars or a spottingscope were used.
All trees were mapped, and each nest was assigned an identification number. In order to determine reuse of old nests and the number of new nests built, the nests that remained intact from the previous year
Primary Dike (1.2 km)
Fig. 3. Map of Geneva reservoir showing the loca- tion of the bhnds and the trees containing cormorant nests.
Great Basin Naturalist
Vol. 37, No. 1
were mapped before the birds returned. Ob- servations of the eggs and young in the nests were made using four ahiminum poles, 1.5 cm in diameter and 1.6 m long, which could be inserted into one another and se- cured by a wing nut. A side-view mirror at- tached to the end pole afforded a view of the nest. When held at chest level, this as- semblage allowed visual access to most nests. Tape was placed at 0.2 m intervals on the poles to measure nest height from ground or water level.
One new and one rebuilt nest were col- lected from the Springville Colony for fau- nal examination in Berlese funnels and for comparative measurements of numbers, av- erage thickness, and length of each piece of nesting material. Invertebrates found in the nests were collected for identification.
Egg length and width were measured us- ing dial calipers accurate to one-twentieth (.05) of a millimeter. Measurements were made to compare eggs of the same clutch, eggs within the colony, and eggs from dif- ferent colonies. Egg volume was determined using the formula V = 0.526 LB^, where L = longitudinal axis and B = transverse axis (Romanoff and Romanoff 1949).
Average clutch size was determined for each colony by including abandoned nests, but not unused nests. A nest was considered abandoned if it contained at least one egg at any time, but showed no signs of in- cubation for one week. Nests that never contained an egg were considered unused.
Hatching success for each breeding pair was determined by dividing the number of hatchlings that survived one day by the to- tal number of eggs laid in the nest, and for the colony by the total number of hatch- lings that survived one day by the total number of eggs laid in the colony.
Fledgling success was determined for the Geneva Colony by dividing the total num- ber of fledglings that left the nest by the number of hatchlings that survived one day.
Three measurements— weight, length of bill, and tarsometatarsus— were made daily of the Geneva hatchlings to determine growth rates. A number indicating nest and order of hatching was written on tape and attached to the leg of each bird for identi-
fication. Weight was determined using three Sargent- Welch dial spring-balances: a 250 gm balance accurate to 2 gm, a 500 gm balance to 5 gm, and a 2000 gm balance to 10 gm. A small linen bag was attached to the balance hook to hold the bird for weight readings. The bill and tarsometa- tarsus lengths were measured with a clear plastic 150 mm rule. The bill was measured from the corner of the mouth to the tip of the upper mandible. The tarsometatarsus was measured from the point of articulation below the ankle to its articulation with the middle toe.
In 1973 trips were made to all recorded nesting sites in the state to determine which colonies still existed and to conduct a cen- sus of Utah's cormorant population. The colony at Bear River Migratory Bird Refuge was visited in order to photograph the use of an alternative nesting habitat and to de- termine location and size. Letters were also sent to managers of selected state bird ref- uges in an attempt to discover new breed- ing colonies.
Results and Discussion
Breeding colonies of cormorants in Utah have been reported in Cache, Rich, Millard, Box Elder, and Davis counties. The only published reference to breeding in Utah County was a brief mention of the area as a "former" nesting site (Woodbury et al. 1949). This paper, then, is the first detailed report of the nesting of the cormorant on Utah Lake, Utah County.
On 20 June 1973 I visited the Newstate Gun Club located 4 km west of Bountiful, Davis County, Utah. An unreported cormo- rant breeding colony was located with nests in Russian olive (Elaeagnus angustifolia) trees along the south side of the Burton Canal. About 15 trees, averaging 8 m high, were scattered for 1 km along the canal. They contained 42 active nests with 66 well-developed young. In addition, 23 im- mature birds were seen swimming in the canal. The trees containing cormorant nests also held a total of 61 Great Blue Heron nests with young. The caretaker of the
March 1977
Mitchell: Cormorant Biology
property reported that the cormorants had nested there since 1962.
Arrival at Nesting Sites.— In 1972 the first birds to return to the Springville Colo- ny were seen on 4 March. Cormorants were first seen at the Bay Colony on 13 March and at the Geneva Colony on 22 March. In 1973 the first cormorants were seen at the Springville Colony on 7 March and on 14 March at the Bay Colony. Three birds, seen on 26 March, were the first to arrive at the Geneva Colony.
Lockerbie (1942) stated that over a peri- od of years the earliest observation dates for cormorants in Utah was 29 March. The earliest that cormorants have been seen at Bear River Refuge is 5 March (Behle 1958). The arrival of cormorants on 4 March 1972 at Springville during this study is the earliest record for Utah.
In 1972 and 1973 the first returning birds encountered inclement weather conditions, and some ice remained on Utah Lake. It is interesting to note that the first birds re- turned to the Bay Colony a week later than the Springville birds. Yet, all of the birds that nested at the Bay and Geneva colonies returned within four weeks, while birds were still arriving at the Springville Colony and building nests in late May. Since the first cormorants to return to Springville are often subjected to extremely poor weather conditions, their early arrival must provide them with some adaptive advantage. It may be that since it is the largest colony, nesting space is at a premium and the first birds to return are assured of the choicest areas. I observed that those birds that did not return to Springville until May in 1973 were forced to build their nests west of the main cormorant colony.
The reason for the time difference in re- turn to each colony is puzzling. The Bay Colony is subjected to essentially the same type of weather conditions as Springville. Geneva may have somewhat different weather, but the pond is always ice-free so the birds could obtain food at any time. The Bay and Geneva colonies are consi- derably smaller in numbers of nesting birds than the Springville Colony, so the lack of
competition for nesting space may not re- quire an early return.
It may also be that the birds in each col- ony remain associated at their wintering grounds and return as a group, or that the individual birds are site specific. Color- banding fledglings to determine if they re- turn to the same colony each year may help answer this question.
Nest Construction.— The Double-crested Cormorant utilizes two different nesting sit- uations determined by local conditions. Of the thirteen colonies reported in this paper, seven were ground-nesting colonies and of these only the Bear River Colony is still in existence. Vermeer (1973) stated that cor- morants prefer to nest on the ground and nested in trees only after their ground nest- ing habitat was disturbed. However, this is not always the case, since the birds at Ge- neva could have nested undisturbed on the dike in association with California gulls, but instead nested in trees.
Most cormorants that nest along the At- lantic coast nest at ground level on islands bare of vegetation (Bent 1922; Lewis 1929; Mendall 1936). Townsend (Bent 1922) found that these island nests consisted of seaweed, kelp, grass, and a few sticks. The nest di- mensions averaged 52 cm across by 12 cm deep.
Unlike the flimsy nests described by Townsend (Bent 1922), Lewis (1929), and Mendall (1936), Behle (1958) found that the ground-nesting cormorants at Egg Island built elaborate interwoven nests of sage and greasewood twigs. The twigs averaged 12 to 18 mm thick and 290 to 580 mm long. The nests were built on or near rocks and some of the nest cups were lined with gull feath- ers. Nests that had been used for more than one season were almost cemented solid with excrement. Cormorants did not nest on the islands at Bear River Refuge until three years after the islands were built (Marshall 1937). At that time rocks were placed at each end of the islands and cormorants then used them for nesting. Yet, Behle (1958), Lewis (1929), Mendall (1936), and Munro (1927) report nesting on bare ground in comparable island situations.
Nest Utilization.— Of the 83 cormorant
6
Great Basin Naturalist
Vol. 37, No. 1
nests located in the Springville Colony in 1973, 14 (17 percent) of the nests left from the previous year were utilized by the first returning birds. In the entire colony only five nests were destroyed and two aban- doned. The height of the nests averaged 5.2 m (range 2.4 to 7.6 m) above the surface of the water. The nests per tree varied from 1 to 11. There did not seem to be any corre- lation between the size of the tree and the number of nests the tree contained. Nests were still being constructed in the eastern part of the colony during late May.
Of the seven cormorant nests studied at the Bay Colony during the nesting season, none were lost, although two nests were abandoned. The nest height averaged 5.3 m (range 4.3 to 6.4 m). Nests per tree varied from 1 to 3. All nest-building ceased by early April.
In 1972 cormorants nested in four trees on the dike at the Geneva Colony (Table 1). Trees 2, 3, and 4 were all being utilized by 1 April, but tree 1 was not used for nesting until about 10 days later (Fig. 3). Eighteen nests from the previous year were reused, although some were in very poor condition. One old nest in tree 1 was not used, although it seemed in good condition. No nests were abandoned. One nest each in trees 2 and 4 were lost when they fell out of the trees during a storm.
The same trees used for nesting at Ge- neva in 1972 were occupied in 1973. How- ever, tree 1 was utilized first in 1973, fol- lowed by trees 2 and 4, and finally by tree 3. Twenty-four nests remained of the 33 used in 1972. Some of these were in very poor condition and only 11 (33 percent) of
Table 1. Summary of nest utilization at the Geneva Colony indicating numbers of new and rebuilt nests in each tree used for nesting. The number in parentheses indicates the percent of total colony nests each tree contained.
|
Tree 1 |
Tree 2 |
Tree 3 |
Tree |
i4 |
|||
|
Year new |
rebuilt i |
new rebuilt i |
new |
rebuilt |
new |
rebuilt |
|
|
1972 1973 |
0 0 |
4(12) 3(18) |
3 6(27) 2 0(12) |
3 2 |
3(18) 1(18) |
9 2 |
5(43) 7(52) |
them were reused for nesting. Nest height averaged 4.7 m (range from 3 to 7 m). Three unused nests were destroyed during bad weather and another was lost due to rob- bing of material by cormorants. Two nests that contained eggs were abandoned before hatching. During both years nest construc- tion was finished by early May.
The reuse of old nests requires a com- ment. Palmer (1962) stated that rebuilding a nest took only two days, compared to four to six days to build a new one. Rebuilt nests were also reported to be much stronger than new nests. Rebuilding generally con- sisted of adding some material to the top of the old nest. He gave one record of a nest being reused for four years. I know of four nests at Geneva which were reused for at least three consecutive years.
At all of the colonies there appeared to be a limit to the proximity of nests, since, even when there was a suitable site for an- other nest, none was built closer than about 1 m to another. The average distance be- tween nests at the Geneva Colony was 1.8 m.
Lewis (1929) described inland nests in Ar- kansas, Minnesota, and Quebec that were located in various species of trees growing on islands or standing in water. The nests were located 0.3 to 22 m from the surface of the ground or water. He reported that the nest dimensions averaged 22 x 48 cm and that there were never more than six nests in one tree. However, Palmer (1962) observed that trees have contained as many as 36 nests and I observed a tree with 11 nests at Springville. The nests consisted of dead twigs, devoid of bark, and the nest cup was often lined with green leaves or grass and gull or cormorant feathers.
I observed that dead twigs from various species of trees were the primary materials used in nest building. However, large pieces of hardstem bullrush were often used. I was not able to identify the tree species used, but I often saw cormorants pick up dead twigs from the ground beneath the same trees they nested in. Large twigs were se- cured in the crotch of branches and then other twigs were interwoven until the nest was approximately 30 cm high and 40 cm
March 1977
Mitchell: Cormorant Biology
wide. Reused nests tend to increase in size each year, due to the addition of material. The nest cup was hned with finer materials, including leaves and cormorant feathers. Occasionally, parts of live and dead bull- rushes were also used to line the nest cup. One new and one rebuilt nest were collect- ed for comparison of building materials. The new nest had materials averaging 312 mm in length (range 150 to 610 mm) and 7.8 mm thick, the thickest being 26.0 mm. The rebuilt nest had materials averaging 333 mm in length (range 170 to 690 mm) and 8.1 mm thick, the thickest being 25.0 mm.
Prior to measuring, the nests were placed in Berlese funnels to collect invertebrates. A beetle {Denneste.s lardarius) and several flies {Rhegoclema sp.) were identified. These in- vertebrates should not be considered a de- finitive list of organisms that are found in company with cormorants. They are only examples of what may be found, as I have encountered no other list of this type in the literature.
From my observations in 1972 I con- cluded that old nests, in good condition, were the preferred nesting habitat of re- turning cormorants. There was an order of preference from those highest in a particu- lar tree to the lowest. Only after these were occupied were new nests built. Despite this general rule, at Geneva in 1973 new nests were built while old nests were left unoccu- pied.
There is evidence that the most mature birds arrive at a colony and begin nesting first (Lack 1968, Palmer 1962). If that were true, the higher nests in a colony may have a better hatching success, since the older birds should be more prolific and make few- er nesting errors (Lack 1968, Lewis 1929). To test this, I took the average of all nest heights for each colony and determined the hatching success for the nests situated above the average height as compared to those be- low the average height. In every instance, the lower nests had a higher hatching suc- cess (Fig. 4). This may indicate that the first returning birds are not necessarily the most mature birds, or possibly that the higher nests, which are more exposed to the ele-
|
7,5 |
-1 |
r |
||
|
7.0 |
- |
|||
|
6.5 |
67* |
' |
r |
|
|
6.0 |
19% |
39% |
||
|
5.5 |
. |
|||
|
I 5.0 |
- |
62* |
||
|
4.0 |
69« |
36% |
||
|
3.5 |
||||
|
3.0 |
- |
|||
|
2.5 |
. |
- |
||
|
2.0 |
Fig. 4. Distribution of nest height showing hatching success above and below the average nest height.
ments when the adults are disturbed (not shielded by leaves or branches), have a higher incidence of pre-hatching mortality. However, more research is needed since neither of these ideas seems a completely adequate explanation.
In Utah it seems to be the rule that cor- morants nest in the same trees with Great Blue Herons. Presently in Utah only the cormorants at Bear River do not nest in company with herons. Former nesting colo- nies in Utah where herons nested with cor- morants were located at Bass Pond (Pearson 1927), Cache Valley (Stanford 1937), and Egg Island (Behle 1958). However, this seems to be an uncommon occurrence throughout most of the cormorants' range, since there are few other references to them nesting with Great Blue Herons. Howell (1911), Stoddard (1922), and Vermeer (1973) reported them nesting with herons in Il- linois, Wisconsin, and Manitoba, respective- Iv. Ganier (1933) described cormorants shar-
Great Basin Naturalist
Vol. 37, No. 1
ing a colony with Great Blue Herons, Water Turkeys, and American Egrets in Tennessee. Cormorants have also been re- ported nesting with Great Blue Herons and Black-crowned Night Herons in Nebraska (Lewis 1931).
I have observed that when cormorants nest in the same trees with herons, the cor- morant nests are always located below the level of the heron nests. Stanford (1937) noted the same thing at the colony in Cache Valley. At Springville and Geneva, where only one heron was nesting in a tree occupied by cormorants, the heron's nest was the topmost in that tree. When heron nests were located at the same level as a cormorant nest, the heron nests were never situated closer than 3 m to it. At Geneva in 1972 the highest nest in tree 1 was utilized by a Great Blue Heron. In 1973 the same nest was used by a cormorant, later desert- ed, and again used by a heron during the same season.
Although there is no explanation in the literature for partitioning of the nesting habitat by these two species, I feel this con- dition may exist for two reasons. Cormo- rants are noted for the large amount of ex- crement they deposit outside the nest, to the point that the very trees they nest in are often killed (Lewis 1929; Mitchell pers. obser.). The feces could also be harmful if they were deposited on heron eggs or young located beneath a cormorant nest. However, the most probable reason is the difference in take-off and landing patterns of the two birds. The cormorant is not an agile bird and, in order to become air- borne, must launch itself precipitously out of the nest (Hall 1926). This may result in eggs or young being kicked out of the nest. Likewise, when they land at the nest, they will sometimes miss their footing and fall out of the tree or crash into their nest-mate or other cormorants nesting nearby. In com- parison, the Great Blue Heron is able to fly away from the nest with considerable grace by launching itself straight into the air, or land at the nest with amazing agility so as to not disturb any of the nesting birds lo- cated nearby. Therefore, I feel that the stratification of the nests of these two spe-
cies is a behavioral adaptation for partition- ing the nesting habitat to avoid interspecific collisions, which could result in the loss of eggs and young.
Further research may answer important questions related to nesting, such as whether cormorants return to the same colony each year or even the same tree and nest. It is not known whether tree-nesting cormorants could effectively become ground nesters. An attempt to induce the ground-nesting cor- morants at Bear River Refuge to nest in trees was successful when old nests were placed in nest boxes in an artificial tree made of 2 x 4's on the island in unit no. 4 (Gunther, pers. comm.). It also needs to be determined whether cormorants could be enticed to a new nesting area by taking old nests from a former colony and placing them in a new location.
Egg Laying and Incubation.— In 1973 egg laying began at the Springville Colony on 26 March and continued into late May. Egg laying began at the Bay Colony about 6 April and continued into early May, and at the Geneva Colony began on 13 April and continued until 17 May.
The difference in the beginning egg lay- ing dates at each colony is a reflection of the time differential in early arrivals at each colony. There is a three-week period be- tween the date of first arrivals and the date that eggs are first laid, so that eggs are laid first at Springville and last at Geneva.
The eggs laid on 26 March 1973 at Springville comprise the earliest record for Utah (Behle 1958), although Lewis (1929) stated that Utah had the earliest laying dates for any state and theorized that laying began during the second half of March. Five records from Utah, given by Townsend (Bent 1922), provided dates of 9 April to 17 May for egg laying. The earliest reported date for eggs laid in Utah was 28 March 1937 at Egg Island, Great Salt Lake (Behle 1958). Marshall (1937) reported a very late egg laying date at Bear River Refuge, where nests contained viable eggs on 2 Sep- tember 1936.
Eggs were laid over a 12-week period at Springville and over more than a 4-week period at Geneva (Fig. 5), although the bulk
March 1977
Mitchell: Cormorant Biology
14 16 18 20 22 24 26 28 30 2 4 6 April
DATE
10 12 14 16 18 20 22 May
Fig. 5. Distribution of egg laying at Geneva Colony from 14 April to 20 May 1973.
of laying at Geneva came within a 2- week period. Vermeer (1969) reported egg laying lasted for a period of 2.5 months in Alberta, Canada. At Geneva 90 percent of the females laid one egg per day until the clutch was complete. The other females took more than a day to lay one or more of their eggs. Others have noted that females laid one egg per day until the clutch was complete (Behle 1958; Lewis 1929; Mendall 1936), although Lack (1968) stated, that cor- morants lay eggs every two to three days.
Lewis (1929) and Palmer (1962) reported that incubation usually began after depo- sition of the third egg. Mendall (1926) and Behle (1958) observed that incubation may begin with the first egg, or at any time thereafter, until the last egg is laid. I found that incubation began immediately after the first egg was laid, which may be due to egg laying beginning early in Utah while the weather is still very cool.
The incubation periods given by Behle (1958), Lewis (1929), and Mendall (1936) were 25 days. Bergtold (1917) reported that most North American cormorants averaged 28 days for incubation, and Worth (1940)
gave the incubation period for the Eu- ropean Cormorant (Phalacrocorax carbo), though somewhat larger in size than P. a. auritus, as 28.5 days. Van Tets (1959), working with the svibspecies P. a. cincinatiis in British Columbia, had 27 records for in- cubation, varying from 25 to 29 days, with most eggs hatching after 28 days. For the 16 Geneva hatchlings in 1973 incubation varied from 26 to 30 days with an average of 28 days for incubation. The difference between the dates reported by others and my results may again be due to early nest- ing and cooler weather.
At Geneva in 1972, 11 nests lost one or more eggs before hatching. All of the eggs from four of the nests were lost before 10 May and a replacement clutch was laid for each by 29 May. The replacement clutch consisted of the same number of eggs as lost in three of the nests and one more egg than lost in the other. In 1973, 12 nests lost one or more eggs prior to hatching. However, of the seven nests that eventually lost all eggs, no clutch was completely lost before 24 May, and no new clutches were laid.
Renesting occurred in 30 to 50 percent of
10
Great Basin Naturalist
Vol. 37, No. 1
cormorants in Manitoba whose eggs or nests were destroyed (McLeod and Bondar 1953). Double-crested Cormorants have never been reported to raise two broods in a single year. It appears that as long as one of the original eggs remains in the clutch, there is no replacement of lost eggs. Also, there is apparently a period during which the fe- male can receive a stimulus that will enable her to lay a new set of eggs. After that time (10 May to 24 May in Utah), she is unable to replace a lost ckitch.
Clutch Size.— The normal clutch size of the Double-crested Cormorant is three or four eggs, with four eggs most common (Bent 1922; Davie 1900; Lewis 1929; Men- dall 1936). Palmer (1962) reported a range of two to seven eggs with one record of nine eggs. Lewis (1929) found that most cormorants along the Gulf of St. Lawrence had a clutch of four eggs, "rarely" five eggs, and 5-10 percent contained three eggs. Mendall (1936) reported the following per- centages for Maine: two eggs, 8 percent; three eggs, 40 percent; four eggs, 50 per- cent; five eggs, 2 percent. In a census taken on Egg Island, Great Salt Lake, in 1941, Behle (1958) found that over half the nests contained four eggs.
I censused eggs weekly in the Springville and Bay colonies and daily in the Geneva Colony; then the maximum number of eggs present was counted before hatching began. A total of 126 eggs were found in the Springville Colony (average 3.8 eggs per nest), and 28 eggs in the Bay Colony (aver- age 4.0 eggs per nest) in 1973. At the Gen- eva Colony in 1972 a total of 80 eggs were found (average 4.2 eggs per nest), and in 1973, 61 eggs (3.6 average) were found. The average for all the colonies in 1972 and 1973 was 3.8 eggs per nest. Table 2 in- dicates the combined totals and percentages of all the clutches from the Utah Lake colo- nies. These figures agree with other find- ings, although there is a greater percentage of clutches containing four and five eggs than previously reported.
£gg Measurements.— Measurements of the longitudinal and transverse axes of the cor- morant eggs indicated that the Bay Colony eggs had the smallest average length and
Table 2. Numbers of eggs laid per nest at the Utah Lake colonies. The percent of the totals are in paren- theses.
|
Year |
Number of |
eggs /nest |
|||||
|
Colony |
2 |
3 |
4 |
5 |
6 |
||
|
Geneva Springville Bay Geneva |
1972 1973 1973 1973 |
0 2 0 1 |
3 6 2 5 |
10 21 3 11 |
5 4 2 0 |
1 0 0 0 |
|
|
Totals |
■3(4) |
16(21) |
45 (59) |
11(14) |
1(1) |
greatest average width of 59.7 x 39.0 (longi- tudinal X transverse in mm), while the Springville eggs showed the greatest aver- age length and smallest average width of 61.35 X 38.5. The Geneva averages were in- termediate in size at 60.43 x 38.99. Small variations from these measurements can be found in the literature as indicated in Table 3, but they are remarkably slight when the large geographic distances separating the samples are considered.
In a discussion on various means of meas- uring eggs, Preston (1968) stated that meas- uring all the eggs in a colony produced a bias toward the larger clutches. If measure- ments are being made in order to establish an average or norm for a species, it is the number of clutches or sets of eggs, not the number of eggs, that is important. He thought that a random selection of one egg from each of 20 or more clutches would be a better indicator for the species. In order to check that hypothesis, I took an average of all the eggs measured at Geneva in 1972 and in 1973. I then compared those aver- ages with three random samples of one egg measured from each of 12 clutches during each year (Table 4). T-tests (two-tailed) at 5 percent indicated that the random means were not significantly diff^erent from the overall mean. However, since Preston rec- ommends that the comparison be done with 20 or more clutches, significant differences may appear in larger samples.
Egg Volumes.— Volumes for the eggs measured at Geneva in 1972 and 1973 were calculated in order to establish an average
March 1977
Mitchell: Cormorant Biology
11
Table 3. Egg measurements from various sources compared with those of the Utah Lake colonies, showing range in length and width measurements with averages in parentheses.
|
Date |
Location |
Sample Size |
Length (mm) |
Width (mm) |
Source |
|
- |
U.S. Nat. Museum" Canada Maine Egg Island, Utah Total Bay Springville Geneva Springville Geneva Total |
40 50 15 20 125 7 7 48 10 52 124 |
56.00-65.42(61.60) 54.50-66.50 (59.90) |
36.50-42.00 (38.80) 34.50-42.00 (37.70) -- (37.60) 34.80-39.40(37.10) |
Bent 1922 Lewis 1929 Mendall 1936 Behle 1958 |
|
- |
53.00-64.00 (59.30) |
||||
|
4/16/71 4/27/71 5/18/72 5/2/73 5/8/73 |
Mean (60.33) 58.06-60.51 (59.70) 58.34-64.05 (61.52) 56.80-63.90 (60.34) 58.08-64.52(61.18) 56.05-65.50 (60.52) |
Mean (37.80) ,38.17-40.38(39.00) ,35.47-41.33 (38.20) .36.90-41.00 (38.93) ,37.17-40.08(38.86) 35.40-40.80 (39.05) |
White, pers. comm. White, pers. comm. This study White, pers. comm. This study |
||
|
Mean (60.65) |
Mean (38.80) |
'These were probably collected from Minnesota, North Dakota, Manitoba, Saskatchewan, and Quebec
for Phalacrocorax auritus, since no informa- tion on egg volumes for this species was re- ported in the hterature at that time. The vokimes were calculated by using the for- mula Volume = 0.526LB2, where L = lon- gitudinal axis and B = transverse axis (Romanoff and Romanoff 1949). The mean volume for the 48 eggs measured at Geneva in 1972 was 48.09 ± 2.80 cc, with a range of 41.55 to 52.69 cc. The mean volume for the 52 eggs measured at Geneva in 1973 was 48.60 ± 3.89 cc, with a range of 38.00 to 54.24 cc.
Table 4. Average length and width measurements of all eggs measured at the Geneva Colony in 1972 and 1973, compared to random samples of ont egg from each of 12 clutches.
Year Sample
Length(mm) Width(mm)
1972 Overall mean 60..34
1972 1st random sample 56.67
1972 2nd random sample 60.,38
1972 3rd random sample 60.69
1973 Overall mean 60.52 1973 1st random sample 60.08 1973 2nd random sample 60.55 1973 3rd random sample 60.67
38.93 ,39.18 38.74 38.94
39.05 ,39.06 .39.26 38.92
Romanoff and Romanoff (1949) have stated that the volume (V) for an egg can also be calculated if the weight (y) is known by the equation V = 0.913 y for precocial birds and V = 0.959 y for altricial birds, such as the cormorant. This seems a very unreliable method, since cormorant eggs have a tendency to lose weight after they have been incubated for a time. Rahn and Ar (1974) reported that during incubation all bird eggs may lose approximately 18 percent of their initial weight. The average weight for 12 unincubated cormorant eggs was reported to be 46.9 gm, while 21 in- cubated eggs averaged 46.2 gm (Lewis 1929).
As a further check on the reliability of this method, I calculated the volumes of 20 eggs, for which the weight was known, and listed the volumes computed for each egg by using both methods (Table 5). The aver- age weight of all the eggs was very close to the average of 46.5 gm for 50 eggs weighed by Lewis (1929). However, there is a differ- ence of almost 4 cc between the average volume calculated from length and width measurements and the average volume cal- culated by weight. A T-test (two-tailed) at 5 percent indicated that there was a signifi- cant variation between these means. Further
12
Great Basin Naturalist
Vol. 37, No. 1
Table 5. A random selection of 20 eggs, taken at different times and from different Utah Lake colonies, show- ing the volume as figured by length and width measurement compared to volume computed from weight. (Ar- ranged according to increasing weight.)
|
Width |
Length |
Weight |
Volume " |
Volume " |
|
|
No. |
(mm) |
(mm) |
(gm) |
(cc) |
(cc) |
|
1 |
35.47 |
58.54 |
37.61 |
.38.59 |
36.07 |
|
2 |
38.33 |
60.10 |
42.03 |
46.44 |
40.30 |
|
3 |
38.72 |
58.08 |
42.45 |
45.80 |
40.71 |
|
4 |
38.17 |
58.06 |
42.58 |
44.60 |
40.84 |
|
5 |
37.62 |
58.34 |
42.62 |
43..35 |
40.87 |
|
6 |
38.99 |
58.87 |
43.39 |
47.12 |
41.61 |
|
7 |
.37.17 |
62.74 |
43.54 |
45.64 |
41.75 |
|
8 |
.39.11 |
59.63 |
44.78 |
47.93 |
42.95 |
|
9 |
37.43 |
59.43 |
44.84 |
43.70 |
43.00 |
|
10 |
38.85 |
59.37 |
44.93 |
47.28 |
43.09 |
|
11 |
38.01 |
64.05 |
45.40 |
48.69 |
43.54 |
|
12 |
38.53 |
61.96 |
45.79 |
48.34 |
43.91 |
|
1.3 |
39.21 |
60.08 |
47.50 |
48.50 |
45.55 |
|
14 |
39.86 |
59.88 |
48.16 |
50.16 |
46.18 |
|
15 |
39.73 |
64.52 |
49.58 |
.53.47 |
47.54 |
|
16 |
38.92 |
62.97 |
49.64 |
50.15 |
47.60 |
|
17 |
41.33 |
61.92 |
50.86 |
55.54 |
48.77 |
|
18 |
39.86 |
61.59 |
51.74 |
51.58 |
49.61 |
|
19 |
40.08 |
60.48 |
52.05 |
51.17 |
49.92 |
|
20 |
29.16 |
61.14 |
52.79 |
49..39 |
50.63 |
|
Means |
38.73 |
60.59 |
46.11 ±4.05 |
47.87 ±3.81 |
44.22 ±3.87 |
' Volume as computed by formula; Volume = 0.526 leugth x breadth^. ' Voliune as computed by formula; Volume = 0.959 x weij^ht.
evidence of the unreliability of using weight to calculate volume is evidenced by com- paring samples #18 and #20 (Table 5). Al- though sample #20 is smaller in both di- mensions, it weighs over 1 gm more than sample #18. Therefore, the formula given by Romanoff and Romanoff (1949) for cal- culating volume from the weight of the egg (V = 0.959 y for altricial birds) may hold true for some species, but is not accurate for the Double-crested Cormorant.
Recently, Pagnalli et al. (1974) reported a mean volume for eight Double-crested Cor- morant eggs from Maine as 45.87 ± 2.75 cc. Their method involves tracing a profile of the egg and dividing the tracing into seg- ments, then calculating the surface area of each segment, and from that the volume. However, they appear to be working with a somewhat larger bird, since they report a mean egg weight of 49.88 ± 3.37 gm, and I found a mean of 46.11 ± 4.05 gm.
The length of the eggs was also different since they reported a mean length of 6.19 ± 0.21 cm and mine were smaller at 6.04 ± 0.19 cm.
Additionally, Preston (1974) has recently stated that the volume of an egg cannot be accurately calculated on the basis of length •and width measurements alone without er- rors of 5 percent or more commonly occur- ring. Because the cormorant egg is pointed at both ends and describes the shape of a bicone, the coefficient given by Romanoff and Romanoff (1949) would be too large. Therefore, my values for volumes would be overestimated and may result in the differ- ence between my data and those of Pagnalli et al. (1974).
Hatching — Hatching at the Springville Colony began on 22 April and continued until the second week in June. At the Bay Colony hatching began on 30 April and ended on 27 May. In 1972 hatching began
March 1977 Mitchell: Cormorant Biology
Table 6. Summary of the egg laying and hatching success for the Utah Lake colonies.
13
|
Colony |
Year |
No. |
Average |
Total No. |
Average |
Total No. |
Hatching |
|
Nests |
Height (m) |
Eggs |
Clutch Size |
Hatched |
Success |
||
|
Geneva |
1972 |
19 |
unk. |
80 |
4.2 |
44 |
55% |
|
Springville |
1973 |
33 |
5.2 |
126 |
.3.8 |
83 |
66% |
|
Bav |
1973 |
7 |
5.3 |
28 |
4.0 |
15 |
54% |
|
Geneva |
1973 |
17 |
4.7 |
61 |
3.6 |
18 |
30% |
at Geneva on 19 May and continued until 10 June. In 1973 hatching began on 20 May and lasted until 9 June. Clutch size, total number of eggs laid, and hatching success are presented in Table 6.
The hatching dates reported for each col- ony reflect an average of 28 days for in- cubation. The earliest hatching began at Springville on 22 April 1973 but, owing to the extended arrivals of new birds, lasted well into June. There was a remarkable syn- chrony of hatching periods at Geneva dur- ing 1972 and 1973. In 1972 hatching began on 19 May and continued for 21 days, while in 1973 hatching began one day later and lasted 20 days (Fig. 6). Two to four days are usually required for the complete hatch- ing of a single clutch since incubation starts soon after the first egg is laid (Mendall 1936). Hatching of the first and last bird may be separated by as much as a week or more (Palmer 1962). When this is the case, the youngest birds do not usually survive.
Ricklefs (1969) stated that the hatching success for most colonial nesting birds .should be over 50 percent. There is no re- port of hatching success for P. a. auritus, but a record for P. a. cincinatus revealed a hatching success of 60.4 percent (Drent et al. 1964). The hatching success at Spring- ville Colony (66 percent) and Bay Colony in 1973 (54 percent), and Geneva Colony in 1972 (55 percent), provided an average of 58 percent (Table 6). That is probably a fair indicator for this subspecies, although the hatching success for grovmd-nesting cormo- rants may vary considerably from that val- ue, due to other external factors. The very low hatching success at Geneva in 1973 (30 percent) requires some explanation other than predation. I believe that the daily vis- its I made to Geneva, in combination with the cool weather during incubation, may principally account for the low success. In order to complete my daily counts, it was sometimes necessary to keep the adults
3 3
^ 2
19 20 21 22 23 24 25 26 27 28 29 30 31 1 2 3 4 5 6 7 May June
DATE
Fig. 6. Distribution of egg hatching at Geneva Colony from 19 May to June 1973.
9 10 11
14
Great Basin Naturalist
Vol. 37, No. 1
from the nest for as long as 45 minutes, al- though the interval was usually less. While awaiting my departure, the adults flew to the reservoir and when I left immediately flew back to the nest. Their moist feathers and water on the eggs may have resulted in an extended period before optimum in- cubation temperature was again reached (Lack 1968).
Ricklefs (1969) has stated that high tem- perature in eggs, caused by exposure to the sun, is a serious mortality factor. This may have contributed to the mortality of some of the eggs in exposed nests, but may not be an important factor, since temperatures were not high until the last of May, when 90 percent of hatching was already com- pleted.
Seven of the Geneva nests that were regularly attended by the adults in 1973 contained only addled eggs and made up 39 percent of all Geneva eggs. That is consid- erably higher than the 24 percent of addled eggs reported for a colony of ground-nest- ing cormorants {P. a. cincinatus) in British Columbia (Drent et al. 1964). All nests at Geneva that contained only addled eggs were still being incubated by the adults two weeks after the expected hatching date. Three of these nests were still being in- cubated 30 days after the estimated hatch- ing date. Incubation of addled or infertile eggs by cormorants beyond the normal in- cubation period is well documented (Snow 1960).
Preckition of Eggs and Ytiing.— Predation on the eggs and young of cormorants is common. Ferry (1909) observed predation by Ring-billed Gulls {L(irits delawarensis) on the eggs of cormorants in Saskatchewan. Lewis (1929) reported predation on eggs and young by Great Black-backed Gulls {Lams tnarinus), crows {Corvus brachy- rhynchos), and ravens (Corvus corax). This only occurred when the adults were absent from the nests, since they are usually in- trepid defenders of their nests. Herring Gulls {Lams argentatus) have also been ob- served attacking eggs and young of un- guarded cormorant nests (Bourget 1973; Mendall 1936). At Mandarte Island in Bri- tish Columbia, crows and gulls removed
eggs from nests that had been vacated for only a few minutes (Drent and Guignet 1961, Van Tets 1959), and 28 percent of the eggs in the colony were lost due to these predations in 1963 (Drent et al. 1964). Cali- fornia Gulls are known to attack and con- sume the eggs and young of unguarded nests (Behle 1958, Odin 1957, Vermeer 1970). The personnel at Bear River Refuge informed me that they had to employ meth- ods to discourage gulls from nesting because they were so destructive to the eggs and young of water birds, including the cormo- rant.
There are no records of mammalian pre- dators on the eggs and young of cormo- rants. Lewis (1929) reported that, even when there were large numbers of coyotes {Canis latrans) in an area where a nesting colony was accessible, they were never known to attempt any attack.
Contrary to these reports and the state- ment made by Armstrong (1965) that gulls always attack eggs in vacated cormorant nests, I never observed mammalian or avian predation at any of the Utah Lake colonies. In addition, I never saw any evidence of predation, such as broken eggs, ravaged young, etc., which would indicate attacks during my absence, although I anticipated serious problems from the California Gulls when I disturbed the nesting birds at Gen- eva. But even when I kept the cormorants from their nests for as long as 45 minutes, 1 never observed an aggressive act toward the eggs or young from a gull, although hun- dreds were flying nearby. It is not sugges- ted, however, that the gulls never attack cormorant eggs and young in this area.
Food Items.— Most of the studies dealing with the food of the Double-crested Cormo- rant have been along the Atlantic coast (Lewis 1929; Mendall 1936). Therefore, they contain lists of marine species of fish and do not provide good information on the food habits of inland cormorants. The few re- cords that do exist include those of the Bio- logical Survey (Lewis 1929) that reported the stomachs of nine birds collected at the mouth of Bear River contained the follow- ing: carp {Cyprinus carpio), 73 percent; Utah Sucker {Catostomus ardens), 19 per-
March 1977
Mitchell: Cormorant Biology
15
cent; Chub {Leuciscus litieatiis), 3 percent; Chub {Leuciscus sp.), 6 percent. Behle (1958) found the following species regur- gitated by young birds at Egg Island, Great Salt Lake: carp, catfish. Yellow Perch (Perca flavescens), Silverside Minnows {Richanl- sonius hydrophlox), and Webug Suckers {Ca- tostomus fecundus). Other records (Lewis 1929, Mendall 1936) have reported cormo- rants eating Bluegill Sunfish (Lepomis pal- lidus). Pumpkin-seed Sunfish {Lepomis gih- bosus), Northern Crappies {Pomoxis sparoides). Common Crappies {Pomoxis an- nularis), Northern Pike {Esox lucius). Wall- eye {Stizostedion vitreum), and a mudpuppy {Necturus maculosus). Palmer (1962) record- ed crayfish, frogs, Fathead Minnows {Pime- phales promelas), and dace {Rhinicthys sp.) as additional food items.
Food items were obtained only from nest- lings that regurgitated when I approached or handled them. Some items that had been partially eaten were taken directly from the nest. Remains of Black Bullhead {Ictahirus melas). Carp, and Utah Chub {Gila atraria) were found at all three colonies; Goldfish {Carassius auratus) were found only at Gen- eva Colony, and White Bass {Roccus chry- sops) only at the Springville Colony. The fish found ranged in size from 3 cm (gold- fish) to 22 cm (bullhead).
Of these five species of fish the nestlings were known to use for food, the Utah Chub, White Bass, and Goldfish are report- ed here for the first time.
Development of the Young.— When the young birds hatched they were naked, blind, and barely able to move, and they had a taut black skin, although Lewis (1929) and Men- dall (1936) reported that they had a dark- brown skin that turned black within a day or two. From about the fourth day after hatching (Table 7), the oldest birds in the clutch were very aggressive. This sometimes resulted in the smaller or younger birds being deprived of food and ultimately dying of starvation (Behle 1958, Hall 1926). At Geneva three birds apparently starved when only a few days old while their siblings grew normally and continued in good health.
Although starvation may be the major
cause of death of fledglings, very young birds are also susceptible to death by expo- sure during hot weather. On hot sunny days it was necessary for the adults to shade the young, and I have observed the young die within minutes when the adults were fright- ened away. Of the 18 young that hatched at Geneva in 1973, 13 survived long enough to leave the nest. After they were a week old, the birds were better able to move into shade and employed gular flutter to cool themselves as indicated by Bartholomew et al. (1968) and Lasiewiski and Snyder (1969).
Table 7. Reactions to human handling by Double- crested Cormorant nestlings during specific age peri- ods.
Age (days)
Reaction
Can barely hold head up; no physical re- sistance or escape attempts; cower and shiver when held; usually make no sound; crawl toward shade when exposed
4-7 Head held erect to peer at intruder;
make a whining sound; cower when hand reached out; attempt to hide be- hind nest-mates; sit quietly when head or eyes are covered.
8-13 Defecate or regurgitate food when han-
dled; peck at intruders; squawk loudly; peck at nest-mate when placed back in the nest; quiet down when head is cov- ered.
14-22 Climb out of nest onto limbs away from
intruder; defecate or regurgitate when approached; use toes and beak to hang onto any object within reach; will not remain quiet even when head covered; peck viciously when handled.
23- Climb frantically out of nest when ap-
proached; often fall or jump out of nest into water; swim well at first attempt.
The eyes opened when the birds were about four to five days old and the egg tooth dropped off at four to seven days, al- though some birds retained the egg tooth for as long as 12 days. A fine black down appeared about the sixth day. Eight days
16
Great Basin Naturalist
Vol. 37, No. 1
later the birds were completely covered by down. When two weeks old, the young were very adroit in the use of beak and feet for grasping. When taken out of the nest, they would grab onto anything with their beak and toes, and it was difficult to dis- lodge them. I observed a four-week-old bird climb 5 m up a tree to the nest from which it had fallen.
Until about two weeks old, the young were fed a semiliquid diet by regurgitation from the adults. Thereafter, whole or par- tially digested fish were brought to the nest. At i 7 to 19 days the flight feathers first ap- peared along the alar tracts, and the light- colored gular pouch became a deeper yel- low. The young were fully feathered at eight weeks and were easily recognizable on the water or in flight by their brownish backs and cream-colored breasts, in contrast to the glossy black of the adults.
Measurements of the Young.— Three measurements of the nestlings in accessible nests at Geneva Colony in 1973 were taken in order to determine growth patterns. Due to the low hatching success, only 16 young were available for measurements and three of those died within four days. Each bird was weighed daily and linear measurements of the beak and tarsometatarsus taken. Fig- ures 7, 8, and 9 contain summaries of these measurements.
The average weight at hatching was 36 gm (Fig. 7), and tarsometatarsal length was 11 mm (Fig. 8). This is close to Lewis's (1929) measurements of 37 gm and 12 mm, respectively. Mendall (1936) reported an av- erage weight at hatching of 32 gm. The av- erage beak length of the Geneva birds at hatching was 17 mm (Fig. 9), but Lewis (1929) reported an average length of 11 mm. The discrepancy is due to different measuring techniques. I measured along the side of the beak from the corner of the mouth to the tip of the upper mandible, and Lewis measured from base to the tip along the top of the beak. The average weight for Mendall's (1936) birds at five days was 158 gm, and mine averaged 156 gm. At 14 days Lewis (1929) reported an average weight of 778 gm, Mendall report- ed 785 gm, and mine averaged 736 gm. At
20 days the Geneva young averaged 1158 gm; Mendall (1936) recorded an average of 1232 gm. The average weight for the Gen- eva young after the 23-day period was 1543 gm (Fig. 7). Gonsiderable weight addition takes place thereafter, since Lewis (1929) gave the average weight for adult males as 2100 gm and adult females as 1670 gm. Those values differ from birds collected in Maine where 10 adult males averaged 2233 gm and 12 adult females averaged 1861 gm (Kury 1968). The Utah birds are even larger than those in Maine. Four specimens taken from Springville in 1974 averaged 2247 gm for two females, and 2553 gm for two males.
Other than the averages recorded at hatching, there is no additional information for beak measurements and tarsometatarsal growth in the literature. When 23 days old, the young had an average beak length of 73 mm and an average tarsometatarsal length of 58 mm. Palmer (1962) records the aver- age tarsometatarsal length of adult birds as 61 mm. Measurements of an immature cor- morant (BYU Life Sciences Museum) in- dicated a beak length of 89 mm and tarso- metatarsal length of 58 mm. The relationship of these measurements to the growth patterns values are indicated in Figs. 8 and 9. The growth curve of the beak indicates continual increase through the 23rd day, but the tarsometatarsal length levels out at approximately 57 mm after the 19th day.
Nest Departure.— There is little agree- ment among authors as to when young cor- morants first leave the nest and how they accomplish their departure. Some, but not all, of the variation can be explained by the difference in nest location, such as nesting on low islands, on cliffs, or in trees. Towns- end (Bent 1922) noted that the young re- mained in tree nests until fully feathered and flew at about eight weeks of age. How- ever, if they fell into water before they could fly, they would make no attempt to swim. Lewis (1929) reported that the young stayed in tree nests until about six weeks old, when they made their first flight. Men- dall (1936) observed that young birds flew feebly to the water from cliff nests at six
March 1977
Mitchell: Cormorant Biology
17
weeks of age, but were not able to dive un- til seven weeks old. Lack (1968) stated that cormorants remained in the nest until they could fly and did not swim until after they were able to fly. Cormorants on Egg Island,
Great Salt Lake, were observed to swim and dive long before they were able to fly, although no age was noted for either event (Behle 1958). In contrast. Burns (1921) found that the young usually fell, rather
|
16 |
16 |
13 |
13 |
13 |
12 |
11 1 |
1 n 10 u |
8 |
|
|
1600 |
|||||||||
|
1500 |
■ |
||||||||
|
1400 |
■ |
||||||||
|
1300 |
|||||||||
|
1200 |
|||||||||
|
1100 |
|||||||||
|
1000 |
.. |
||||||||
|
900 800 |
t |
||||||||
|
700 |
|||||||||
|
600 500 |
+ |
||||||||
|
400 300 200 |
+ |
f |
f |
||||||
|
100 |
-♦- |
+ |
+ |
DAY
Fig. 7. Mean increase in weight (gm) from measurements on alternate days. The vertical line represents the range, the horizontal line the mean, and the blackened area is equal to two standard deviations. The sample size is given at the top of each day's measurements.
18
Great Basin Naturalist
Vol. 37, No. 1
than flew, into the water from nests in trees. Some of these birds were only 17 days old when they fell from the nests.
I observed that very few tree-nesting cor- morants in Utah remained in the nest until they were able to fly. After the young were approximately three weeks old, they would leave the nest and move onto tree limbs in
order to increase the distance between them and me, which often resulted in young birds falling into the water. At first I suspected that they had drowned, but I later observed them swimming as far as 15 m underwater. After a large number had fallen out of the nest they could be seen swimming in groups some distance from the colony. I presumed
Fig. 8. Mean increase in tarsometatarsal length (mm) from measurements on alternate days. The vertical line represents the range, the horizontal line the mean, and the blackened area is equal to two standard deviations. The sample size is given at the top of each day's measurements.
March 1977
Mitchell: Cormorant Biology
19
that my presence caused early departures until I visited the colony at Newstate Gun Club in June 1973. At the time of my visit the birds had not been disturbed for three weeks, yet when I arrived at the colony there were 23 immature birds between
three and six weeks of age swimming in the canal. I satisfied myself that they were not able to fly by chasing them up and down the canal. While I was there, other young fell into the water and two birds that fell onto the ground immediately walked over
Fig. 9. Mean increase in beak length (mm) from measurements on alternate days. The vertical line represents the range, the horizontal line the mean, and the blackened area is equal to two standard deviations. The sample size is given at the top of each day's measurements.
20
Great Basin Naturalist
Vol. 37, No. 1
to the canal and swam away. Early depar- ture from the nest would require the young to learn quickly to fish or to continue to be fed by the adults (Behle 1958, Lewis 1929, Mendall 1936). I have never observed young cormorants being fed on the water, but I assume the adults continue to feed them until they are able to fish effectively for themselves.
The dates stated below pertain to the time when the birds were sufficiently devel- oped to be absent from the nest and prob- ably survive. This does not imply that the birds were able to fly when they left the nest. In all observed cases the birds fell or jumped from the nest into the water and swam expertly away. This made it impos- sible to estimate the date when the fledg- lings were able to fly. The earliest date at which birds were absent from the nest at Geneva in 1973 was 17 June, and the last nestling left on 5 July. Age at the time of leaving was between three and six weeks. The oldest bird to leave was 37 days old and the youngest was 21 days old. The av- erage age for leaving the nest was 29 days.
Migration.— Most of the local cormorants leave the Utah Lake area in October. Al- though I have seen them on Utah Lake as late as 18 November, there is no way to tell if these are local birds or others migrating through. The latest date they have been seen at Bear River is 1 December (Behle 1958). Palmer (1962) stated that almost all of the Atlantic coast migrants have arrived at their wintering grounds in Texas and Louisiana by 1 December.
An unusual sighting was made by Reed Ferris in 1959 (Scott 1959) when he saw a cormorant at Geneva reservoir on 1 Janu- ary. Since the reservoir is supplied with warm water, it does not freeze during the winter and the bird would be able to fish. However, there was no way to tell whether the bird was a late southern migrant, a bird that had overwintered, a very early arrival, or an injured bird.
Discussion
Status of Utah Lake Colonies.— Provo Bay has been a major nesting site for cor-
morants for over 50 years (Bee 1920), al- though it has received little attention. The population there has at times exceeded 300 nesting birds (Bee 1949). Smaller numbers of cormorants have nested in the Geneva Colony but have probably done so contin- uously for almost 30 years (Bee 1946). The total number of cormorants breeding in the Utah Lake area for the past five years has probably remained stable at around 100 nesting pairs, although the numbers utilizing each colony fluctuate (Table 8).
Table 8. Numbers of active nests in the Utah Lake colonies over a five-year period.
|
Year |
Spriri |
.gvi |
ille |
Bay |
Geneva |
Totals |
|
1970 |
Unk. |
14 |
26 |
Unk. |
||
|
1971 |
68 |
17 |
18 |
103 |
||
|
1972 |
48 |
12 |
31 |
91 |
||
|
1973 |
83 |
6 |
15 |
104 |
||
|
1974 |
96 |
7 |
0 |
103 |
Utah's Cormorant Population.— In 1958 seven cormorant breeding colonies were re- ported in Utah, but five had been deserted or destroyed (Behle 1958). Behle made no mention of the colonies that were located on White Rock, Great Salt Lake, and Rock Island, Utah Lake, or the existing Utah Lake colonies (Fig. 2). However, the former two sites were limited to a few birds and were probably used only briefly.
In 1974 the only Utah cormorant colonies in use were located at New State, Utah Lake, and Bear River Migratory Bird Ref- uge. The latter consists of two groups of birds nesting on two small islands, one lo- cated in unit four and the other in unit five. Although there is no published reference to nesting since 1949 (Behle 1958), records kept by the refuge personnel indicate con- tinuous usage through 1974.
The population of Double-crested Cormo- rants in Utah has been steadily decreasing for the past 50 years. Behle (1935) reported a decrease in cormorant numbers on Egg Is- land from 500 birds in 1915 to 114 in 1935. He felt that the breeding cormorants in the
March 1977
Mitchell: Cormorant Biology
21
region were threatened with extinction. Sugden (1936) noted an alarming decrease in the numbers of herons and cormorants nesting on the Great Salt Lake and made a plea for their protection.
Of the 13 colonies that have existed at one time or another within the state, only five were still in use in 1974 (Table 9). Dur- ing the 1930s and 1940s the larger colonies may have easily supported a combined total population of well over 1000 birds. How- ever, in 1973 the total known cormorant population of Utah consisted of only 386 birds nesting in five colonies (Table 10).
Table 9. Summary of the populations of Utah's cor- morant colonies and their current status.
Table 10. Summary of Utah's population as of 1973.
existing cormorant
|
Largest |
recorded |
||||
|
no. nesting |
1973 |
||||
|
Colony |
birds |
Year |
status |
||
|
Bear River |
600 |
1947 |
94 |
||
|
New State |
84 |
1973 |
84 |
||
|
Geneva |
62 |
1972 |
30 |
||
|
Springville |
166 |
1973 |
166 |
||
|
Bay |
300 |
1949 |
12 |
||
|
Bass Pond |
80 |
1927 |
destroyed |
||
|
Bear Lake |
400 |
1929 |
deserted |
||
|
Logan |
400 |
1939 |
deserted |
||
|
Dolphin Island |
150 |
1919 |
deserted |
||
|
Egg Island |
500 |
1915 |
deserted |
||
|
Farmington |
Bay |
°2 |
1951 |
deserted |
|
|
Rock Island |
°8 |
1937 |
deserted |
||
|
White Rock |
°2 |
1901 |
deserted |
||
|
"Thesf are the a |
luthor'^ |
, estimates base |
don |
the hteratiire. |
Colony
Percent of No. nesting birds state total
|
Springville |
166 |
43 |
|
Bear River |
94 |
24 |
|
New State |
84 |
22 |
|
Geneva |
30 |
8 |
|
Bay |
12 |
3 |
|
Totals |
386 |
100 |
for years. Entire colonies have been wiped out and their numbers reduced by the thou- sands (Buchheister 1944, Demille 1926, Lewis 1929, McLeod and Bondar 1953, Smith 1911, Stanford 1937). The diking and subsequent draining of Provo Bay will only compound an already critical situation for Utah's steadily declining Double-crested Cormorant population unless adequate measures are taken to protect their present nesting sites.
Acknowledgments.— I appreciate the en- couragement and suggestions offered by Dr. Herbert H. Frost during my research and for his critique of this manuscript. I thank Dr. Clayton M. White for his suggestions and critique of this manuscript and Dr. Stephen L. Wood for the identification of insects found in the nests. Appreciation is also extended to the U.S. Steel Company, Geneva Works, for allowing access to the Geneva Colony.
Effects of Diking Provo Bay — Two of the five remaining cormorant colonies in Utah are located on Provo Bay, Utah Lake (Fig. 2). In 1973 the Springville Colony was the largest in the state and supported 43 per- cent of the total remaining birds in Utah and, combined with the small Bay Colony, comprised 178 birds or 46 percent of the state's total cormorant population (Table 10). The proposal to dike and drain Provo Bay as part of the Central Utah Project (Fig. 2) would eliminate the nesting area of these birds.
Cormorants have been subjected to per- secution in the United States and Canada
References
Armstrong, E. \. 1965. Bird display and behavior. Dover Publications, Inc., New York.
Bartholomew, C. A., D. C. Lasiewiski, and E. C. Crawford. 1968. Patterns of panting and gular flutter in cormorants, pelicans, owls, and doves. Condor 70:31-34.
Bee, J. 1946. Unpublished field notes. University of Kansas, Lawrence, Kansas.
Bee, R. G. 1920. Bee Oological Collection. Life Sci- ences Museum, Brigham Young University, Pro- vo, Utah.
1937. Unpublished field notes, 3:105. Life Sci- ences Museum, Brigham Young University, Pro- vo, Utah.
22
Great Basin Naturalist
Vol. 37, No. 1
1949. Unpublished field notes, 6:93-106. Life
Sciences Museum, Brigham Young University, Prove, Utah.
Behle, W. H. 1935. A history of the bird colonies of the Great Salt Lake. Condor 37:24-35.
1958. The bird life of Great Salt Lake. Univ.
Press, Salt Lake.
Bent, A. C. 1922. Life histories of North American petrels and pelicans, and their allies. U.S. Nat. Mus. Bull. 121:236-261.
Bergtold, W. H. 1917. A study of the incubation pe- riods of birds. Kendrick-Bellamy Co., Denver.
BoiRGET, A. A. 1973. Relation of eiders and gulls nest- ing in mixed colonies in Penobscot Bay, Maine. Auk 90:809-820.
BucHHEisTER, C. W. 1944. The comeback of the cor- morants. Audubon 46:14-29.
Burns, F. L. 1921. Comparative period of nesting life of some North American Nidicolae. Wilson Bull. 33:90-99.
CoTT.\M, C. 1927. Distributional list of the birds of Utah. Unpublished master's thesis, Dept. of Zo- ology, Brigham Young University, Provo, Utah.
D.-vviE, O. 1900. Nests and eggs of North American birds. David McKay, Philadelphia.
Demille, J. B. 1926. Birds of Gaspe County, Quebec. Auk 43:510-514.
Dre.nt, R. H. and C. J. GuiGNET. 1961. A catalogue of British Columbia sea-bird colonies. Occas. Pa- pers Brit. Columbia Prov. Mus., No. 12.
Drent, R. H., G. F. Van Tets, F. Tampa, and K. Vermeer. 1964. The breeding birds of Mandarte Island, British Columbia. Can. Field Nat. 78:208-263.
Ferry, J. J. 1909. Birds observed in Saskatchewan dur- ing the summer of 1909. Auk 27:189-190.
Ganier, a. F. 19.33. Water birds of Reelfoot Lake. J. Tenn. Acad. Sci. 8:65-83.
Gu.nther, L. 1974. Manager of Bear River Migratory Bird Refuge. Personal communication.
Hall, R. E. 1926. Notes on water birds nesting at Pyramid Lake, Nevada. Condor 28:87-91.
Howell^ A. H. 1911. Birds of Arkansas. U.S. Biol. Surv. Bull. 38:15-16.
KuRY, C. H. 1968. Differences in weight of male and female cormorants. Auk 85:113.
Lack, D. 1968. Ecological adaptations for breeding birds. Methuen and Co. Ltd., London.
Lasiewiski, R. C. and G. K. Snyder. 1969. Response to high temperature in nestling Double-crested Cormorants and Pelagic Cormorants. Auk 86:529-540.
Lewis, H. F. 1929. The natural ni. .ory of the Double- crested Cormorant {Phaiacrocorax aurittii aur- itu.s Lesson). Ru-.Mi-Lou Books, Ottawa.
1931. .additional information concerning the
Double-crested Cormorai t (Phaiacrocorax aur- itus auritus Lesson). Auk 1 .5:207-214.
Lockerbie, C. W. 1942. Bird ,■•. about the Great Salt Lake. News Bull. Mi^ieralogical Soc. Utah, 3:fv4-66.
Marshall, W. H. 1937. Double-crested Cormorant nesting on the Bear River Refuge in Utah. Con- dor 39:36.
McLeod, J. A. and G. F. Bondar. 1953. A brief study of the Double-crested Cormorant on Lake Win- nepegosis. Can. Field Nat. 67:1-11.
Mendall, H. L. 1936. The home life and economic status of the Double-crested Cormorant {Phaia- crocorax auritus auritus Lesson). Univ. Maine Studies, Second Ser., No. 38, Maine Bull. 39(3).
Munro, J. A. 1927. Observations on the Double- crested Cormorant (Phaiacrocorax auritus) on Lake Manitoba. Can. Field Nat. 41:102-108.
Odin, C. R. 1957. California gull predation on water- fowl. Auk 74:185-202.
Pagnalli, C. v., a. Olzowka, and A. Ar. 1974. The avian egg: surface area, volume, and density. Condor 76:319-325.
Palmer, R. S. 1962. Handbook of North American birds I. Yale Univ. Press, New Haven.
Pearson, T. G. 1927. A-birding in Utah. Bird-Lore 29:379-383.
Preston, F. W. 1968. The shape of birds' eggs: mathe- matical aspects. Auk 85:454-463.
Preston, F. W. 1974. The volume of an egg. Auk 91:132-1,38.
Rahn, H. and a. Ar. 1974. The avian egg: incubation time and water loss. Condor 76:147-152.
Ricklefs, R. E. 1969. An analysis of nesting mortality in birds. Smithsonian Contr. Zoology, No. 9.
Romanoff, A. L. and A. J. Romanoff. 1949. The avian egg. Wiley and Son, New York.
Scott, O. K. 1959. Great Basin, Central Rocky Moun- tain Region. Audubon Field Notes 13 (3):311- 312.
Smith, F. 1911. Double-crested Cormorants breeding in central Illinois. Auk 28:16.
Snow, B. 1960. The breeding biology of the Shag (Phaiacrocorax aristotelis) on the Island of Lun- dy, Bristol Channel. Ibis 102:554-575.
Stanford, J. S. 1937. Cormorant and heron colonies in Cache Valley, Utah. Proc. Utah .\cad. Sci., Arts, Letters 14:195.
Stoddard, H. L. 1922. Bird notes from southern Wis- consin. Wilson Bull. 24:67-71.
Sugden, J. W. 1936. Sanctuaries of the bird islands of Great Salt Lake. Auk 53:289-294.
Van Tets, G. F. 1959. A comparative study of the re- productive behavior and natural history of three sympatric species of cormorants (Phaiacrocorax auritus. Ph. penicillatus. Ph. pelogicus) at Man- darte Island, B. C. Unpub'ished master's thesis, Univ. British Columbia
Vermeer, K. 1969. Coloni ; oi ' . .hlc •.• ' Cormo-
rants and White Pi .i. .i" e { n. Field Nat. 83:.36-39.
1970. Some aspe. - " h nes'; .. Double- crested Cormorant., a, ^^, Saskatche- wan, in 1969; a plea t. r ,)roteccion. Blue Jay 28:11-13.
1973. Great Blue Heron and Double-crested
March 1977 Mitchell: Cormorant Biology 23
Cormorant colonies in the prairie provinces. 1949. Annotated checklist of the birds of Utah.
Can. Field Nat. 87:427-4.32. Bull Univ. Utah 39(16) Bio. Ser, 11(2): 1-40.
White, C. Professor of zoology, Brigham Young Uni- Worth, C. B. 1940. Egg volume and incubation peri-
versity, Provo, Utah. Personal communication. od. Auk 57:44-60. Woodbury, A. M., C. Cottam, and J. W. Sugden.
CATTAIL DECLINE AT FARMINGTON BAY WATERFOWL MANAGEMENT AREA'
Christine Minton Anderson'
Abstract.— For the last seven years, 1969 to 1976, a steady decline of cattails (Typha latifolia L.) has been ob- served at Farmington Bay Waterfowl Management Area, Davis County, Utah. Several parameters of Farmington Bay's environment that could cause or enhance a decline of the marshland vegetation were studied. These pa- rameters included temperature, pH, phenols, oil and grease, heavy metals, fungus pathogens, and salt water in- trusion from the Great Salt Lake. Elevated soluble salt concentrations were found to be responsible for the de- cline. Cattails are weakened or killed when they are exposed to soluble salt concentrations greater than 5.0 gm /liter, and there is a significant (1 percent level) negative correlation (-0.68) between soluble salt concentra- tion and cattail height. Symptoms of elevated salt concentrations include stunted growth, leaf tip necrosis (burn- ing), and occasional browning of an entire cattail clone. Furthermore, those salt concentrations that cause phys- iological stress in cattails also facilitate the growth of a decomposition fungus, Chaetophoma confluens. This fungus causes a rot consisting of irregularly scattered lesions on the surface of the rhizomes and was consistently isolated from rhizomes of declining plants.
For the last seven years, 1969 to 1976, a steady decline of cattails has been observed at Farmington Bay Waterfowl Management Area, Davis County, Utah. The management area is 4,205 feet above sea level, located on the east shore of the Great Salt Lake, 15 miles north of Salt Lake City. The primary water source is the Jordan River. Water flows northwesterly, and Farmington Bay flows directly into the Great Salt Lake.
At Farmington Bay, cattails in many areas were less than 1 m high, as compared to healthy cattails that frequently attained heights of 2 m. During the summer, cattails appeared to grow normally until July; growth then ceased. Yellowing and necrosis followed rapidly. Older parts of the cattail clones died first, with the younger shoots gradually following the same pattern. In- spection of the rhizomes revealed the pres- ence of mahogany-colored cankers, or le- sions, scattered irregularly over the surface. In some cases the lesions had extended deep into the tissues, and some rhizomes had completely rotted away.
These symptoms are characteristic of those caused by .soil fungus pathogens, and it was suspected that the lesions might be the result of such a fungus and, in turn, that
the lesions might cause the plant's decline. A fungus might act either as a primary pathogen or a weak, secondary pathogen, affecting tissues subjected first to some other stress. If this were the case, it would be necessary to establish what stresses might be involved and to define their relative contri- bution to the cattail decline.
The objectives of this study were to: 1) determine the cause or causes of the cattail decline; 2) evaluate the seriousness of the situation; and 3) provide information so that corrective or control measures could be de- veloped to protect existing marshland vege- tation and reestablish vegetation that had been destroyed.
A fourth objective was to better under- stand the marsh ecosystem. In this system, decomposition is an ecological factor that may be important in affecting marsh plant interactions. The declining cattails offered an opportunity to study how fresh-water marshlands might undergo decomposition and change.
The broad-leaved cattail {Typha latifolia L.) is common to North American marsh- lands and dominant in Utah's waterfowl management areas. Other common marsh
'This article is based on research that was done by the author for a master's thesis. The research was partially funded by the .\ssociated Students i the University of Utah's Environmental Studies Program.
'Biology Department, University of Utah, Salt Lake City, Utah 84n2.
24
March 1977
Anderson: Cattail Decline
25
plants are alkali bulrush (Scirpus paludosus A. Nels.), hardstem bulrush (Scirf)us acutiis iVIuhl.), salt grass {Distichli.s stricta Torr. Rydb.), and salicornia (Salicornia rubra A. Nels. and Salicornia pacifica Standi.).
Important as a successional species, the cattail is a plant that traps sediment and nutrients, allowing the subsequent estab- lishment of bulrushes. Bulrushes, the climax species of the marsh, provide abundant wa- terfowl food. Cattails, on the other hand, have little value as waterfowl food, and, when it is dominant in waterfowl manage- ment areas, the cattail is considered a nui- sance because its dense stands make it use- less as nesting habitat or escape cover. Cattail rhizomes are palatable to humans, muskrats, and geese; the plants themselves are used by waterfowl for nesting and es- cape cover when their growth has been controlled.
The most significant publication on the subject of Utah marshlands contains the re- sults of 25 years of experimental work at Ogden Bay Refuge, 12 miles west of Ogden, Utah, near the east shore of the Great Salt Lake (Nelson 1954). The purpose of the study was to determine how marshland could be developed and waterfowl popu- lations increased; records were kept to as- certain how plant community changes in- fluenced animal populations. Several of the conclusions were pertinent to the present study at Farmington Bay.
Major factors limiting plant growth and waterfowl use at Ogden Bay included water supply, seasonal water fluctuations, fluctua- tions of the Great Salt Lake, and over- grazing. Salt tolerance differed among plant species. Of the emergent species, alkali bul- rush was the most tolerant; cattail was sec- ond. All of the marsh plants were able to tolerate soil salinities exceeding the salinity ranges for agricultural crops. As plant cover was increased, waterfowl populations in- creased because new nesting areas were cre- ated and abundant food was made available.
After vegetation had been developed on the Ogden Bay Refuge, it became evident that cattails, especially in the less saline areas of the marsh, could effectively out- compete other plants. Dense stands, useless
for waterfowl nesting, were established. In 1947, cattail control studies were initiated by the Utah State Department of Fish and Game (Nelson and Dietz 1966). The control methods used to experimentally eliminate cattails included cutting, crushing, and mowing; chemical sprays; explosives; fire; and drought. It was concluded that by ad- justing the water levels cattail overgrowth could be controlled; when mud flats were not covered with water during the growing season, large-scale cattail invasions occurred.
At Fish Springs, a salt marsh west of Salt Lake City, Typha angustifolia (narrow- leaved cattail) does not form large commu- nities (Bolen 1964). Bolen noted that Scirpus acutus and Typha latifolia, common species of Utah's fresh-water marshes, did not occur at Fish Springs. He concluded from his data that while it was probable that salinity lim- ited the development of these species, water depth or some other factor was probably re- sponsible for their absence at Fish Springs.
In contrast to the Fish Springs study, data on physical and chemical soil constituents at Bear River Bird Refuge indicated that so- luble salts affected the yields of aquatic plants. Areas low in soluble salts were more productive than areas high in soluble salts (Jensen 1940). It was determined that high salinity, in addition to affecting vegetative growth, interferes with seed germination and seed production of Utah marsh plants (Kaushik 1963).
Literature published on other localities stresses the importance of soil salinity in in- teractions between marsh plants (Haller 1974). Salinity is considered to be a primary factor in determining plant distributions in Louisiana marshes; cattails are limited to fresh-water areas of less than 0.5 percent salt (Penfound and Hathaway 1938). Water levels, in addition to salinity, are also thought to determine plant distribution (Bourn and Cottam 1939).
Cattail populations growing in disturbed .salt flats have salt tolerances characteristic of each species. A concentration of 1 per- cent sodium chloride in hydroponic tanks caused stunting, drying, and leaf curl in Typha latifoha. Typha angustifoUa was able to withstand 2 percent sodium chloride con-
26
Great Basin Naturalist
Vol. 37, No.
centrations. A hybrid intermediate, Tijpha glauca, had an intermediate salt tolerance (McMillan 1959). Since many terrestrial plants also differ widely in their salt toler- ance (Levitt 1972), an effort was made to determine the relationships between cattail decline and salinity at Farmington Bay.
Methods
Five different features of the problem were studied: 1) the isolation of a possible fungus pathogen; 2) the investigation of en- vironmental stress conditions; 3) the eval- uation of marshland by aerial photography; 4) laboratory experimentation on the effects of salinity on cattail growth; and 5) labora- tory experimentation on the effects of sali- nitv on the growth of an isolated fungus.
1. Isolation of the fungus: Diseased cat- tails were collected, and plant tissue was cut from necrotic lesions with a sterile scal- pel. The infected tissue was surface ster- ilized and placed on Difco-Bacto Agar with nitrogen or Difco potato dextrose agar. Plant tissue was surface sterilized with a 10 percent aqueous solution of Purex bleach.
Pure cultures were obtained by transfer- ring the developing fungus mycelium onto fresh agar. Cultures were allowed to grow at room temperature. Slant test tube cul- tures provided a method of storage, and they were refrigerated until used. The fungi were identified. Fungus mycelium was mounted on temporary slides with a drop of water for most identifications. To aid in the identification of one species {Chaetophoma confluens), fixing, sectioning, and staining were used. Tissue was fixed in formalin- aceto-alcohol and embedded in paraffin. Sections, 15 microns thick, were stained with safranin 0 according to the methods described by Johansen (1940).
2. Environmental parameters: Studies were conducted to determine the environ- ment's contribution to the vegetation de- cline. This included measuring salinity, pH, biological oxygen demand (BOD), dissolved oxygen (DO), and temperature. Samples of soil and water were taken weekly during the months of May through September, 1974. All routine sampling was done at the
four sampling sites shown on Figure 1. Cat- tails appear to be healthy at Site 1 and Site 3. Cattails are in a state of decline at Site 2; Site 4 represents the outlet of Farm- ington Bay and has no cattails.
Conductivity of both soil extracts and wa- ter samples was measured in the laboratory at 25 C with a conductivity bridge (Rich- ards 1954). Then pH was measured to the nearest tenth of a pH unit with a Beckman pH meter. BOD samples were taken once a month and analyzed by the Utah State Di- vision of Health. Water samples for dis- solved oxygen were measured weekly in the field according to the azide modification; DO was measured in milligrams per liter and expressed relative to 100 percent satu- rated water (American Public Health Asso- ciation 1971). In addition, samples were chemically analyzed monthly by the Utah State Division of Health to determine which industrial pollutants were present. Air and water temperature were measured in the field with a centigrade thermometer. The data obtained were used to calculate the percent of dissolved oxygen in the wa- ter. Weather conditions for the summer were recorded and compared with previous summer conditions.
3. Mapping of vegetation: Aerial photo- graphs and moving films of Farmington Bay and the surrounding area were taken. The Ogden Bay Waterfowl Management Area, the Bear River Bird Refuge, and Red Butte Canyon were also examined for indications of vegetation decline in order to determine if the decline was in some way uniquely as- sociated with Farmington Bay. A vegetation map of Farmington Bay was made so that the affected area could be properly illus- trated and the extent of damage could be determined. A U-2 infrared aerial photo- graph (scale 1:120,000) was projected onto a United States Geological Survey Map (Farmington Quadrangle, scale 1:24,000) with a Bausch and Lomb Zoom Transfer Scope. Areas covered by marsh vegetation were traced onto the geological survey map. The U-2 infrared photograph, obtain- ed from the Geography Department at the University of Utah, was taken 14 September 1972 by a high-flying aircraft at about
March 1977
Anderson: Cattail Decline
27
5(),()()() feet. Standard infrared film was used in a high-resolution camera. Healthy vegeta- tion is bright red, vegetation under stress is salmon colored, and dead vegetation is white. A compensating polar planimeter was used to determine the area covered by healthy marsh vegetation and the extent of diseased vegetation.
Color aerial photography for assessing the quality of marshlands has just recently come into use. Further information on the meth-
ods used in aerial mapping of marshlands can be obtained from Seher and Tueller (1973); the source also explains how infrared photographs should be interpreted. In addi- tion, a bibliography on the use of aerial photography (Anson 1975) has been com- piled that provides more detailed informa- tion.
4. Laboratory experiments determining ef- fects of salinity on cattail growth: Cattails were grown from seeds gathered at Farm-
Healthy vegetation Stressed vegetation
'ying vegetation
The dashed line shows where vegetation was in 1969.
Fig. 1. Farmington Bay Waterfowl Management Area.
28
Great Basin Naturalist
Vol. 37, No. 1
ington Bay. Seedlings were transferred to six hydroponic tanks when they were abont six inches high; the tanks contained a four-salt nutrient solution made according to the methods described by Arnon and Hoagland (1940). Nutrients were replaced monthly during the experiment. Each of six hydro- ponic tanks contained 10 individual plants. Two tanks served as the control; the other four tanks were used to determine the ef- fects of different salt concentrations on the cattails. After the seedlings had acclimatized and had shown vigorous growth, sodium chloride was added in small amounts (0.01 Mole per day) over a period of 30 days. This was done to determine a realistic salt tolerance for the cattails without putting them under undue stress. Shoot growth of each plant was measured every two days. The final concentrations at the end of the period of additions were two tanks with 0.0 M sodium chloride, two tanks with 0.1 M sodium chloride, and two tanks with 0.3 M sodium chloride. The cattails in one tank of each salt concentration were inoculated with Chaetophoma confluens, the fungus that was most often isolated from the le- sions on the cattail rhizomes. Inoculation was performed by injecting a fungus-water suspension into the host with a hypodermic syringe.
Cattail rhizomes were observed through- out the experiment for the development of lesions and other symptoms of stress. When the experiment was terminated, a wet weight for shoots and rhizomes of each plant was obtained. Tissue from all experi- mental plants was cultured for isolation of the fungi that were present.
5. Effect of salinity on the fimgus, Chaeto- phoma confluens: Chaetophoma confluens was grown on potato dextrose agar, to which sodium chloride had been added. The following sodium chloride concentrations were prepared: 0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, and 0.8 M. Cultures of each salt concentration were inoculated with 0.5 cmsquares of mycelium and were allowed to grow for two weeks in an oven held at 30 C. The area covered by fungus myce- lium was determined by tracing the area onto paper, cutting, and weighing. The av-
erage growth area at each salt concentra- tion was graphed.
Results
Three fungus species were isolated from diseased cattails that had been collected at Farmington Bay. Fusarium tricinctum was isolated in 12 (16.2 percent of the isolated fungi) out of 74 fungus isolations. Botrytis cinerea was isolated 14 times (18.9 percent). Both of these fungi are saprophytic on decaying plant material. F. tricinctum oc- curs frequently on moldy corn and grain; B. cinerea causes "gray mold" of numerous or- namental and crop plants (Barron 1968). Neither of these fungi has been reported to occur on Tijpha.
Chaetophoma confluens was isolated 48 times (64.9 percent). Chaetophoma might be an important decomposition fungus in the marsh. It has been isolated from Spartina (cordgrass), a common emergent in salt-wa- ter marshes (Barnett 1956).
Another fungus, tentatively identified as a species of Olpidium, was discovered when the stained sections of cattail rhizomes were examined; the relationship of this fungus to cattail rhizome rot is unclear. Only a few species of Olpidium are parasitic on higher plants. When Olpidium infects clover or vetch, it causes distortions of the leaves; in one species of grass, Agrostis stolonifera, it causes root hair distortion. Chytridiomi/cetes (the fungus class that includes Olpidium) are characterized by the production of motile cells and are typically found in aquatic hab- itats (Sparrow 1943).'
The most probable cause of the vegeta- tion decline was revealed when the results of the water and soil analyses were exam- ined. Figures 2 and 3 show the results of the conductivity measurements. The figures show that the salinity increases dramatically during the summer at Site 2, reaching a maximum during the month of July. This peak coincides with the period during which cattail deaths were first observed dur- ing the summer. The salinity peak also came just after the Great Salt Lake reached its high of 4,201.30 feet above sea level in June (United States Geological Survey
March 1977
Anderson: Cattail Decline
29
1975). The 4,205-foot level is shown along the east side of Farmington Bay on a United States Geological Survey Map of the area; the west side of the management area is approximately 4,200 feet above sea level. Apparently, the management area is at the same elevation as are the Great Salt Lake brines for part of the summer. Chemical analyses of the water at Farmington Bay showed that sodium and chloride ions were present in the greatest amounts at this time. It was concluded that these ions made the largest contribution to the measured con- ductivity values.
To determine the significance of the rela- tionship between salinity and cattail de- cline, another experiment was conducted. This experiment consisted of taking water samples every tenth of a mile (0.16 km) along the east side of Farmington Bay and along the dike that runs east and west across the refuge. As water samples were
taken, the average height of the cattails was recorded in centimeters. Cattails were meas- ured in a circular area of 10 m in diameter. The point where the water sample was taken served as the center of the circle. Av- erage heights were based on measurements of 20 to 50 cattail plants. Only living cat- tails were measured. Figure 4 shows a graph of conductivity versus cattail height. There is a moderately high, negative corre- lation (r = -.68 with 41° of freedom) be- tween conductivity and cattail height. These field data also showed that cattails exhibit disease symptoms when the dissolved salt content exceeds 5.0 gm/1.
The pH of Farmington Bay lies between the normal ranges of river water: 5.0 to 9.0. The data showed that there was no correla- tion between pH and cattail height.
The BOD at Site 2 progressively in- creased during the summer, while the BOD at Site .3 and Site 4 increased slightly and
May
May
< Site #1 ASite #2
I Days
Osite #3 QSite #4
< Site #1 ^Site #2
July August
Time in Days
<>Site 13 j^Site #4
Fig. 2. Conductivity of the water.
Fig. 3. Conductivity of the soil.
30
Great Basin Naturalist
Vol. 37, No. 1
then leveled off (Fig. 5). The large increase indicated that oxidizable organic material was accumulating at Site 2. Results of the DO measurements are shown in Figure 6. Dissolved oxygen at Site 2 got as low as 46 percent saturated. Despite some periods of low DO and high BOD, the dissolved oxy- gen was high overall during the summer and no detrimental effects were observed that could be related to low oxygen avail- ability.
The results of the chemical analyses showed that the chemical parameters of the water were normal except for the high sali- nity. Small amounts of phenols, oil, and grease were found in the water, and a dis- cussion of their significance to this study is included in the Discussion section of this article.
The effect of sodium chloride on the growth of the fungus, Chaetophoma con- fluens, is shown in Figure 7. The addition
of sodium chloride was found to increase growth. Optimum concentration for growth of Chaetophoma is 0.30 M sodium chloride. Cattails growing in hydroponic tanks showed symptoms of salt toxicity (leaf burn and wilting) at sodium chloride concentra- tions of 4.68 gm/1 (0.08 M). At sodium chloride concentrations of 17.6 gm/1 (0.30 M), growth was completely arrested and rhizomes began to rot. The results correlate well with the results of the field study. Table 1 shows how increased salt concentra- tions affected the wet weights of shoots and rhizomes.
Attempts to isolate Chaetophoma from le- sions that developed on cattails following inoculation were unsuccessful. Because of the observed effects of salinity on the cat- tails, and because Chaetophoma was only isolated in two (out of 20) cultures in pre- vious inoculation experiments, it could not be concluded that the fungus was a primary
Conductivity (micromhos/cm at 25° C) Fig. 4. Conductivity vs. cattail height.
May June July August September
Time in Days A Site #2 OSite #3
(^Site #4
Fig. 5. Biological oxygen demand.
March 1977
Anderson: Cattail Decline
31
cause of the cattail decline. The com- bination of the appropriate soluble salt con- centration and the fungus appears to be critical in producing the symptoms observed at Farmington Bay. Figure 8 shows that the optimum sodium chloride concentration for Chaetophoma growth (0.30 M) corresponds to the salt concentration (0.30 M) that causes cattail death.
Table 1. Wet weights of cattail shoots and rhizomes in relation to salinity.'
|
Sodium |
Shoot : |
||
|
Chloride |
Rhizome |
Shoot |
Rhizome |
|
Concentration |
Weight |
Weight |
Ratio |
|
0.0 M |
73.0 gm |
209.7 gm |
2.90 : 1 |
|
0.1 M |
53.4 gm |
58.6 gm |
1.10: 1 |
|
0.3 M |
26.2 gm |
U.lgm |
0.42 : 1 |
These are the average weiijhts of all the plants tha Hiven sodium chloride concentration.
Discussion
At present, the vegetation in Farmington Bay and the neighboring duck clubs appears to be under stress. The stress conditions oc- cur throughout the marsh, leaving very few places unaffected. Healthy vegetation is found only in the higher areas of the marsh and where fresh water is flowing in from the Jordan River.
The portion of Farmington Bay illu.strated in Figure 1 has an area of 11.88 square kil- ometers (2,934 acres). Of this, 4.88 km^ (1,205 acres) is emergent vegetation. In 1969, Farmington Bay had 5.92 km^ (1,462 acres) of emergent vegetation; 1.04 km^ (257 acres) of this vegetation has dis- appeared, and 1.28 km^ (316 acres) of vege- tation is dying and in the process of dis- appearing. It is interesting to note that in 1959 there were only 1.40 km^ (346 acres) of vegetation in the portion of Farmington Bay that is illustrated. At the present time.
<3 Site #1 A5ite #2
Fig. 6. Dissolved oxygen.
• 1 i I ;
1 I / ' "^A^^ I
. 1 0.2 0.3 O.i. 0.5 0.
Sodium Chloride Concentrati
0.7 0.8 0.9
{moles/liter)
Fig. 7. The effects of sodium chloride on the growth of Chaetophoma confluens.
32
Great Basin Naturalist
Vol. 37, No. 1
there is more vegetation in Farmington Bay than there was in 1959. The same general pattern of vegetation decline is occurring in the southern portion of Farmington Bay and in the marshland to the south that is not shown in the illustration. The decline ap- pears to be occurring slowly, and it would take several years of similar conditions to eradicate the emergent marsh vegetation.
Several parameters of Farmington Bay's environment could cause or enhance a de- cline of the marshland vegetation. Among these are adverse temperature ranges, pH, phenols, oil and grease, fungus pathogens, and salt water intrusion from the Great Salt Lake. All of these potential pathogens were examined in this study. The effects that each agent or factor might have on the veg- etation, alone or in combination, were con- sidered.
Since no unusual temperature changes have been recorded at Farmington Bay, temperature can be immediately eliminated as the cause of vegetation decline. Marshes
Sodium Chloride Concentration (moles/liter)
Fig. 8. The effects of sodium chloride on the growth of cattails and Chaetophoma confluens.
in Red Butte Canyon that experience a sim- ilar range of air and water temperatures have shown no signs of diseased or dying vegetation. However, the cattails at Farm- ington Bay are dying during the warmest part of the summer, and it is possible that high temperatures might be enhancing the effects of some other stress to which the cattails have already been subjected.
Soil and water pH greatly affect the availability of plant nutrients. The ideal pH range for most crop plants is from 6.5 to 7.5 (Sprague 1964). Average values above 8.0 or below 5.0 would be necessary to cre- ate serious nutrient deficiencies and cause disease. A pH range of 5.0 to 9.0, with an average pH around 8.0, was measured at Farmington Bay. Chlorosis (yellowing of the leaves), which is indicative of many nutrient deficiencies, and other characteristic symp- toms that commonly occur in response to the lack of essential elements, were not ob- served (Treshow 1970). In addition, no sta- tistical correlation could be obtained be- tween pH and cattail growth as determined by height. In other studies conducted on Utah marsh plants, it has been shown that vegetation grows well in the pH ranges that occur at Farmington Bay (Jensen 1940, Nel- son 1954, Kaushik 1963). There is little pos- sibility, therefore, that pH is contributing to the vegetation decline.
Analysis of the water at Farmington Bay for possible industrial pollutants revealed the presence of phenols, oil, and grease. Phenolic compounds in water resources are usually a result of pollution from oil refi- neries. Phenols also result from the break- down products of herbicides such as 2,4- dichlorophenoxyacetic acid (2,4-D) and 2, 4,5-trichlorophenoxyacetic acid (2,4,5-T) (Goerlitz and Brown 1972). The oil refi- neries upstream from Farmington Bay and the herbicide treatment of some grain crops with 2,4-D are the most probable sources of the phenols, oil, and grease in Farmington Bay. Phenol concentrations ranged from none to 0.008 mg/1. These concentrations are extremely dilute when compared to the concentrations of phenolic compounds (such as 2,4-D) that are known to be toxic to broad-leaved dicots (Crafts and Bobbins
March 1977
Anderson: Cattail Decline
33
1962). Since the concentrations of phenols are dilute, there is little possibility that phe- nols are causing the vegetation decline. However, phenolic compounds might be contributing to the decline by inhibiting the reestablishment of cattail seedlings. McNaughton (1968) has shown that phenols produced by Typha latifolia have an auto- toxic effect on seed germination and seed- ling growth. Oil and grease measurements ranged from 0.6 mg/1 to 8.0 mg/1. These concentrations are also too dilute to cause any vegetation damage (Goerlitz and Brown 1972).
Study of the total salt content, or salinity, of the soil and water of Farmington Bay re- vealed that the existing salinities were in a range that is known to be toxic to many plant species (Richards 1954). Field studies showed that cattails are weakened or killed when they are exposed to soluble salt con- centrations greater than 5.0 gm/1. Penfound and Hathaway (1938) have also found that Typha latifolia plants in Louisiana marshes do not occur where salinities are above 5.0 gm/1. McMillan (1959) found that T. lati- folia grown in hydroponic tanks could not tolerate 10 gm/1, and Kaushik (1963) found that cattails from Ogden Bay Waterfowl Management Area were killed when sub- jected to a soluble salt concentration of 0.9 gm/1. Symptoms of the elevated salt con- centrations on cattails include stunted growth, leaf tip necrosis (burning), and oc- casional browning of the entire plant. Fur- thermore, the salt concentrations that cause a physiological stress to cattails facilitate the growth of a decomposition fungus, Chaetophoma confluens. This fungus causes a rot consisting of irregularly scattered le- sions on the surface of the rhizomes and was consistently isolated from rhizomes of declining plants. At the beginning of this study, it was thought that the fungus might be a primary cause of the decline. How- ever, elevated soluble salt concentration ap- pears to be the primary cause.
The source of the elevated salt concentra- tion at Farmington Bay is undoubtedly the Great Salt Lake. Salinities in Farmington Bay increase and decrease following the ris- ing and falling of the lake. Salinity measure-
ments of the water entering Farmington Bay from the Jordan River show concentra- tions of salt between 0.1 gm/1 and 0.6 gm/1. These concentrations do not seem to be increasing the salinity of Farmington Bay. The Great Salt Lake, on the other hand, with a salt concentration of 224 gm/1, is a natural source of salt for bordering marshlands because the brines are known to diffuse through and under the dikes around the marshes. The periodic fluctuations of the Great Salt Lake allow for a natural con- trol of the density of cattail stands. This could be either beneficial or detrimental, depending on the frequency and extent of the fluctuations. In order to maintain a de- sirable stand density, it is suggested that the salinity be monitored and controlled.
Literature Cited
American Public Health Association. 1971. Stand- ard methods for the examination of water and wastewater. American Pubhc Health Associ- ation, Inc., New York. 874 pp.
Anson, A. 1975. Photogranimetry as a science and as a tool (an index). Photogramm. Eng. 41(2):225- 236.
Arnon, D. I. AND D. R. HoAGLAND. 1940. Crop pro- duction in artificial solutions and in soils with special reference to factors influencing yields and absorption of inorganic nutrients. Soil Sci. 50:4&3-484.
Barnett, H. L. 1956. Illustrated genera of imperfect fungi. Burgess Publishing Co., Minneapolis, Minnesota. 218 pp.
Barron, G. L. 1968. The genera of hyphomycetes from soil. Williams and Wilkins Co., Baltimore, Maryland. .364 pp.
Bolen, E. G. 1964. Plant ecology of spring-fed salt marshes in western Utah. Ecol. Monogr. .34(2): 143-166.
Bourn, W. S. and C. Cottam. 1939. The effect of lowering water levels on marsh wildlife. Trans. North Am. Wildl. Conf. 4:343-350.
Crafts, A. S. and W. W. Robbi.ns. 1962. Weed con- trol. McGraw-Hill Book Co., New York. 660 pp.
Goerlitz, D. F. and E. Brown. 1972. Methods for analysis of organic substances in water. Tech- niques of water-resources investigations of the U.S.G.S., Book 5, Chapter A3. U.S. Government Printing Office, Washington, D.C. 40 pp.
Haller, W. T. 1974. Effects of salinity on growth of several aquatic macrophytes. Ecology 55(4):891- 894.
Jensen, G. H. 1940. The relation of some physical and chemical factors of the soil to the productivity and distribution of certain waterfowl food
34
Great Basin Naturalist
Vol. 37, No. 1
plants at the Bear River Migratory Waterfowl Refuge. Master's thesis, Utah State University- 30 pp.
JoHANSEN, D. A. 1940. Plant microtechnique. McGraw- Hill Book Co., New York. 523 pp.
Kaushik, D. K. 1963. The influence of salinity on the growth and reproduction of mansh plants. Ph.D. dissertation, Utah State University. 123 pp.
Levitt, J. 1972. Responses of plants to environmental stresses. Academic Press, New York. 697 pp.
McMillan, C. 1959. Salt tolerance within a Typha population. Am. J. Bot. 46(7):521-526.
McNaughton, S. J. 1968. Autotoxic feedback in rela- tion to germination and seedling growth in Typha kitifolia. Ecology 49(2):367-369.
Nelson, N. F. 1954. Factors in the development and restoration of waterfowl habitat at Ogden Bay Refuge. Utah State Department of Fish and Game, Salt Lake City, Utah. 87 pp.
Nelson, N. F. and R. H. Dietz. 1966. Cattail control methods in Utah. Utah State Department of Fish and Game, Salt Lake City, Utah. 31 pp.
Penfound, W. T. and E. S. Hathaway. 19,38. Plant communities in the marshlands of southeastern Louisiana. Ecol. Monogr. 8(l):l-56.
Richards, L. A. 1954. Diagnosis and improvement of saline and alkali soils. U.S. Department of Agri- culture Handbook No. 60, U.S. Government Printing Office, Washington, D.C. 160 pp.
Seher, S. J. and p. T. Tueller. 1973. Color aerial photos for marshland. Photogramm. Eng. 39(5):489-499.
Snedecor, G. W. 1957. Statistical methods. Iowa State College Press, Ames, Iowa. 534 pp.
Sparrow, F. K. 1943. Aquatic phycomycetes. Univer- sity of Michigan Press, Ann Arbor, Michigan. 785 pp.
Sprague, H. B. 1964. Hunger signs in crops. David McKay Co., New York. 461 pp.
Treshow, M. 1970. Environment and plant response. McGraw-Hill Book Co., New York. 422 pp.
United States Geological Survey. 1975. Water re- sources data for Utah. Part I, Surface Water Records. U.S.G.S. Water Resources Division, Salt Lake City, Utah. 505 pp.
OSTEOLOGY AND MYOLOGY OF THE HEAD AND NECK REGIONS
OF CALLISAURUS, COPHOSAURUS, HOLBROOKIA, AND
UMA (REPTILIA: IGUANIDAE)
Douglas C. Cox' and Wilmer W. Tanner^
Abstract.- A detailed study of the anterior osteology and myology of Callisatmis, Copho.saurus, Holbrookia, and Vma reveals the phylogenetic relationships among the sand lizards. An SPSS discriminant analysis of os- teological characters combined with myological characters indicates that Callimiirtis is most primitive, Cop- hosaurus and Holbrookia are most closely related, and Uma is the most distinct of the sand lizard genera. Be- cause of close relationships between Cophosaurus and Holbrookia, it is postulated that earlessness evolved once, and Cophosaurus is returned to synonymy under Holbrookia.
Blainville (1835) wrote the first descrip- tion of a sand lizard and named it Calli- saurus draconoides. Since then various au- thors have pubhshed articles concerning sand lizards. Girard (1851) described Hol- brookia maciilata, Trochel (1852) described Cophosaurus texanus, and that same year Baird and Girard synonymized Cophosaurus with Holbrookia, providing the name Hol- brookia texana, which then remained un- changed for over 100 years. Subsequently, Baird (1858) described Uma notata.
By 1858 all genera represented in the sand lizard group had been described. Since then new species and subspecies, as well as new combinations, have been added by var- ious authors as follows: Bocourt (1874) Hol- brookia elegans- Cope (1880, 1883, 1894, 1895, 1896, 1900) Holbrookia lacerata, Hol- brookia maculata fkwilenta, Uma scoparia, Uma inornata, Uma rufopuncatata, Calli- saurus crinitus, Callisaurus rhodostictus, Holbrookia maculata maculata, Callisaurus draconoides ventralis, and Callisaurus ven- tralis gabbii; Stejneger (1890) Holbrookia maculata approximans and Holbrookia maculata lacerata; Richardson (1915) Calli- saurus ventralis myurus; Dikerson (1919) Callisaurus carmenensis; Schmidt (1921, 1922) Holbrookia maculata campi, Hol- brookia pulchra, Holbrookia dickersonae, and Callisaurus ventralis inusitatus; Schmidt and Bogert (1947) Uma exsul; Barbour (1921) Holbrookia thermophila; Harper
'Department of Zoology, Brigham Young University, Provo, Utah 84602. 'Life Science Museum, Brigham Young University, Provo, Utah 84602.
(1932) Holbrookia propinqua stonei; Smith (1935, 1943, 1946:137, 145) Holbrookia ele- gans thermophila, Holbrookia elegans ele- gans, Holbrookia bunkeri, Holbrookia macu- lata ruthveni, Holbrookia maculata dickersonae, Holbrookia maculata pulchra, and Holbrookia maculata thermophila; Lins- dale (1940) Callisaurus draconoides myurus and Callisaurus draconoides gabbii; Heifetz (1941) Uma notata notata; Bogert and Dor- som (1942) Callisaurus draconoides brevipes; Smith and Burger (1950) Holbrookia prop- inqua propinqua and Holbrookia propinqua piperata; Peters (1951) Holbrookia texana texana and Holbrookia texana scitula; Axtell (1956) Holbrookia lacerata, Holbrookia lacer- ata subcaudalis and Holbrookia maculata perspicua; Smith and Cochran (1956) Calli- saurus draconoides rhodostictus; and Wil- liams, et al. (1959) Uma paraphygus.
Some summaries, reviews, checklists, and comparative studies have also been written. Cope (1896) synonymized Uina and Calli- saurus in a short paper di.scussing the genus Callisaurus. He recognized Uma again in his large work on the crocodilians, lizards, and snakes of North America (1900) and recognized one species and three subspecies of Callisaurus.
Smith (1946:137, 145), in his "Handbook of Lizards," recognized in Callisaurus one species and ten subspecies and stated (p. 145): 'The whole group of Callisaurus of
35
36
Great Basin Naturalist
Vol. 37, No. 1
western United States is in need of revision. The subspecies are not adequately charac- terized, nor are their ranges well worked out. There is very little information on the life history." Although we now know much more about the life history of CaUisaurus, it is still in need of a comprehensive tax- onomic review and remains a monotypic genus. (A careful revision for CaUisaurus has not yet been attempted.)
The genus Holbrookia has had two revi- sions. Schmidt (1922) made the first; a sec- ond was by Axtell (1958). Schmidt (p. 709) stated: "The taxonomy of the North Ameri- can lizards of the genus Holbrookia Girard offers one of the most interesting and diffi- cult problems in North American herpeto- logy."
He recognized several species and states that (p. 712) ^'Holbrookia is obviously di- rectly related to CaUisaurus, from which it differs only in the concealed tympanum and with which it agrees in general features of color pattern and scutellation."
Smith (1946:137, 145) doubts that a prac- tical means of characterizing the species ex- ists and believes that "until such a means is found there will remain indefinitely a prob- lem in defining the ranges of the several subspecies, or in defending their actual validity. Accordingly the genus, par- ticularly the maculata group, merits a care- ful study perhaps more than any other in the United States."
Axtell (1958) recognized only three spe- cies, HoUjrookia lacerata, H. propinqua, and H. maculata. He considered H. texana to be a species of the genus CaUisaurus and therefore did not discuss it. He did list two subspecies for lacerata, two for propinqua, and 11 for maculata.
Uma has been reviewed by Heifetz (1941), Norris (1958), and Mayhew (1964b). Heifetz recognized U. notata notata, U. no- tata cowlesi, U. inornata, and U. scoparia. He also referred to the taxonomic confusion that exists because of erroneous type local- ities in this genus. In his checklist Schmidt (1953) recognized only one species (notata), with three subspecies {notata, inornata, and scoparia).
Norris (1958) recognized Uma n. notata.
U. n. rufopunctata, U. n. inornata, U. sco- paria, and U. exsul. He discussed the evolu- tion of Uma and its relationship to other sand lizards.
The conflict relative to the classification of the notata-scoparia group was discussed by Mayhew (1964b). He recognized U. in- ornata, U. notata, and U. scoparia all as full species on the basis of temperature toler- ance and reproductive data.
Peters (1951) reviewed Holbrookia texana {Cophosaurus texanus). He described two sub- species, but mentions little concerning rela- tionships with other sand lizards.
Clarke (1965) revived Cophosaurus texan- us on the basis of behavioral data collected in a large comparative study of the sand fizard group.
Ecological and behavioral studies con- cerning sand lizards have also been pub- lished by Burt (1931a, 1931b), Stebbins (1944, 1954, 1966), Ramsey (1948, 1949), Cagle (1950), Williams and Smith (1958), Axtell (1960), Lannom (1962), Carpenter (1963, 1967), Clarke (1965), Mayhew (1964a, 1964b, 1966), Pianka and Parker (1972), Tanner and Krogh (1975), and Judd (1974, 1975).
The anatomy of these lizards has not been thoroughly studied. Earle, in a series of articles (1961a, 1961b, 1961c, 1962), de- scribed in detail the comparative anatomy of the middle ear of sand lizards. Stebbins (1943, 1944) described the nasal structures and some aspects of the ecology of Uma, then (1948) described the nasal structures of lizards in general, which included the sand lizards. Axtell (1958) described the osteology of Holbrookia and stated that it is essentially the same as found in all sand lizards. Eth- ridge (1964) studied the skeletal morphology of sceloporine lizards, which includes sand lizards, and compared their relationships. Savage (1958) studied Urosaurus and Uta and made remarks concerning related gen- era, which included sand lizards. A few ref- erences to sand lizards were made by Lar- sen and Tanner (1974) while studying Sceloporus, and Guttman (1970b) also refers to them in his electrophoretic study of the hemoglobins of sand lizards. He found that
March 1977
Cox: Lizard Morphology
37
all genera possessed the same major and mi- nor protein components.
Only portions of the osteology have been adequately treated; the myology is essen- tially untouched. The objectives of this study are to: (1) describe the skull osteology and branchiomeric myology of sand lizards; (2) identify osteological and myological characteristics that distinguish the sand liz- ard generic groups; and (3) determine more accurately the relationships between these groups. No attempt will be made to deal with the species and subspecies except as they relate to the generic phylogeny.
Materials and Methods
Specimens used were obtained from the Brigham Young University Life Sciences Museum (BYU), the California Academy of Sciences (CAS), and the Strecker Museum at Baylor University. At least four speci- mens from each of the four genera were used for osteological examinations. The myology of these 16 specimens, plus four additional individuals from each genus, were also used in developing the myological description. The following list indicates the material used: Callisaums draconoides gabhi from N.T.S., Nye Co., Nevada: (BYU) 2943, 2967, 3079, 40037; C. d. inusatatus from Tiburon Island, Sonora, Mexico: (BYU) 30175, 30176, 30178; C. d. splendidus from Isla Angel de la Guardia, Gulf of California, Mexico: (BYU) 41112; C. d. carmenesis from Baja California Sur, Mexico: (BYU) 41095, 41231; Cophosaunis texanus texanus from Chihuahua, Mexico: (BYU) 14339, 15712; C. t. scitula from Sierra Co., New Mexico: (BYU) 30512, 30513, 30515; C. t. scitula from Pima Co., Arizona: (BYU) 34331, 34336; Holbrookia lacerata from Giaraz Co., Texas: (CAS) 73979; H. maciilata approx- imans from Colonia Dublan, Mexico: (BYU) 11370, 17099; H. maculata hunkeri from Chihuahua, Mexico: (BYU) 15782, 15785, 15788, 15789; H. propinqua propinqua from Padre Island, Cameron Co., Texas: (CAS) 16187; Uma notata inornata from Riverside Co., California: (BYU) 3263, 3266, (CAS) 22824, 22826; U. n. cowlesi from Sonora, Mexico: (BYU) 30144, 30156; U. scoparia
from San Bernardino Co., California: (BYU) 11389, (CAS) 42072.
Skeletal material was prepared by careful dissection. Bones were cleaned with forceps and dissecting needles and soaked in Clorox bleach for several minutes to loosen soft tis- sues, after which hirther picking and clean- ing was done. Skulls were not allowed to dry, but were preserved in 70 percent EtOH to insure that cartilaginous skeletal elements could be examined.
The Statistical Package for the Social Sci- ences (SPSS) discriminant analysis was uti- lized to aid in the identification of charac- ters where they were not already obvious. The statistical analysis will be described be- low.
Myological examinations consisted of careful dissection in which each muscle was separated and its origin and insertion deter- mined. General morphology (shape, relative size, and position) of each muscle was also noted. Muscle comparisons are based on both origin-insertion and muscle morpho- logy. Only the branchiomeric muscles asso- ciated with the hyoid arch and the jaws are described.
Osteology
Sand lizard skulls have been examined in detail. The length and width of individual elements were measured with a Golgau ver- nier caliper and a five-millimeter mini-tool. Comparisons were made on the basis of ra- tios, as well as on the shape and position of each bone in relation to other articulating bones. The lower jaw and hyoid have been studied in the same manner.
The skull is streptostylic, with a freely movable quadrate bone. In mature individ- uals the ethmoid region of the braincase proper is not ossified, but consists of carti- lage plates from which cartilaginous rods extend dorsad to add to and support the membranes that protect the brain. Eyes are large, and only a thin sheet of cartilage sep- arates them medially; there is no apparent area of ossification, even in mature individ- uals.
The sand lizard skull follows the basic plan of Sceloporine lizards as described by
38
Great Basin Naturalist
Vol. 37, No. 1
Ethridge (1964) and Larsen and Tanner (1974, 1975); there is a great deal of unifor- mity within the sand lizard group. We will not give detailed descriptions of each bone here; however, these data are available to those who may desire them. We will, how- ever, describe the variations from the Sce- loporine skull as well as the differences be- tween the sand lizard genera.
The following is a list of the skeletal ele- ments measured and described (they are il- lustrated on Figs. 1, 2, 3, 4):
Basioccipital
Supraoccipital
Vomer
Maxilla
Frontal
Parietal wing
Quadrate
Surangular
Coronoid
Orbit
Suprateinporal
Basisphenoid
Pterygoid
Palatine
Nasal
Infraorbital fossa
fossa
Jugal
Postorbital
Dentary
Splenial
Hyoid
Posttemporal fossa
Exoccipital
Ectopterygoid
Premaxilla
Prefrontal
Parietal
Squamosal
Articular
Angular
Temporal fossa
The sand lizard skulls differ from the gen- eral sceloporine skull in that they lack the lacrimal and the postfrontal bones. They differ among themselves in several ways, but most of the variance seems to be associ- ated with the posterolateral position of the skull. The discriminant analysis selected 116 ratios that could aid in discriminating the four genera. Six measurements were found to be important. Every ratio selected in- cluded at least one of these six. They are listed along with their frequency of use: squamosal width (26), quadrate length (26), mandible length (20), hyoid length (20), mandible width (19), and maxillarv length (15).
Representative ratio comparisons are giv- en to illustrate the separation and relation- ships of genera. Figures (1-4) of the lateral, dorsal, and ventral aspects of the skull are also presented, including the lateral and me- dial views of the lower jaw and the dorsal view of the hyoid.
Earl (1961a, 1961b, 1961c, and 1962) re- ported on the osteological variations in the ear and indicated that all four groups were
distinct, with Cophusaurus midway between CalUsaunis and Holhrookia.
One of the first variations noticed, as skulls were being prepared, was that Co- phosaurus and Holhrookia had a para-occipit- al process directed anterolaterally. That of CaUisaurus and Lhna is directed lateral or slightly posterior. As a result of this varia- tion, variations in the length and width of the quadrate, squamosal, jugal, and postor- bital were noted. In lhna and CaUisaurus the ventral articulating process is ventral to the basioccipital, whereas in Cophosaurus and Holhrookia the quadrate is much short- er. Uma and CaUisaurus also have a wider squamosal, and the squamosal and jugal just meet. In the earless genera the jugal pushes in between the squamosal and postorbital. In one Uma individual the squamosal and jugal failed to meet, and the postorbital had pushed in between them. Some other variations (such as the configuration of su- tures between the frontal and parietal, the frontal and nasals, the basisphenoid and ba- sioccipital, and the maxilla and premaxilla, as well as proportional variations in the maxilla, postorbital, and premaxilla) were also found, but they were slight and had no significance. These are attributed to individ- ual variation on the species or subspecies level.
Myology
The myology of the neck and throat re- gion has been carefully studied. Axtell (1958) pointed out that the inter- mandibularis muscle in the most anterior re- gion of the throat is degenerate. The gen- ioglossus is therefore the most ventral muscle in that region. This condition is con- stant throughout the sand lizard genera. The variations within the sand lizard genera are found in the intermandibularis muscles, depressor mandibularis, constrictor colli, and the omohyoideus-sternohyoideus complex.
The intermandihularis anterior super- ficiaUs (Figs. 5, 6) is a small muscle. In lhna, it is distinctly fan shaped. In CaUi- saurus and Holhrookia it is not fan shaped, and Cophosaurus shows a slight amount of fanning. In Holhrookia the fibers of this
March 1977
Cox, Tanner: Lizard Morphology
39
Fig. 1. The skull osteology of Callisaurus BYU 3079: A. lateral view of skull; B. lateral view of lower jaw; C. medial view of lower jaw; D. dorsal view of skull; E. ventral view of skull; F. dorsal view of hyoid. The symbols used are:
|
an-angular |
ch-ceratohyal |
ju-jugal |
pr-parietal |
|
ar-articular |
co-columella |
mx-maxilla |
pt-pterygoid |
|
bh-basihyal |
cr-coronoid |
na-nasal |
qu-quadrate |
|
bo-basioccipital |
de-dentary |
pa-palatine |
sp-splenial |
|
bp-basipterygoid |
ec-ectoptorygoid |
pf-prefrontal |
sq-squamosal |
|
bs-basisphenoid |
fr-frontal |
pm-premaxilla |
sr-surangular |
|
cbl-ceratobronchlal I |
gh-glossohyal |
po-post orbital |
vo-vomer |
|
cbll-ceratobronchial II |
hh-hypohyal |
40
Great Basin Naturalist
Vol. 37, No. 1
Fig. 2. The skull osteology of Uma BYU 3266: A; B; C; D; E; F; same as Fig. 1. The symbols used are the same as Fig. 1.
March 1977
Cox, Tanner: Lizard Morphology
41
Fig. 3. The skull osteology of Holbrookia BYU 15783: A; B; C; D; E; F; same as Fig. 1. The symbols used
the same as Fig. 1
42
Great Basin Naturalist
Vol. 37, No. 1
Fig. 4. The skull osteology of Cophosaurus BYU 30518: A; B; C; D; E; F; same as Fig. 1. The symbols used are the same as Fig. 1.
March 1977
Cox, Tanner: Lizard Morphology
43
Fig 5 Lateral view of head and neck musculature; superficial depth. A. Lhna BYU 3263; B. Callisaurus BYU 2943- C. Holbwokia BYU 15782; D. Cophosaurus BYU 30512. The symbols used are: am-adductor mandibulans externus medius; as-adductor mandibularis externus superficialis; au-auditory meatus; cc-constrictor colli; dm- depressor mandibularis; gg-genioglossus; iap-intermandibularis anterior profundus; ias-mtermandibulans anterior superficialis; ip-intermandibularis posterior; la-levator angularis oris; mhl-mandibulohyoideus I.
44
Great Basin Naturalist
Vol. 37, No. 1
las iap
99 mh II
mh I pm
d m
.cc sh I sh II oh
Fig. 6. Ventral view of throat musculature; superficial layer at left and first depth at right. A. Uma BYU 3263; B. Callisauru-s BYU 2943; C. Ilolbrookia BYU 15782; D. Cophosaurus BYU 30512. The symbols used are: cc-con- strictor colli; dm-depressor mandibularis; gg-genioglossus; iap-intermandibularis anterior profimdus; ias-mter- mandibularis anterior superficialis; ip-intermandibularis posterior; mhl-mandibulohyoideus I; mhll-mandibulo- hvoideus II; oh-omohyoideus; prn-pterygomandibularis; shi-sternohyoideus I; shil-stemohyoideus II.
March 1977
Cox, Tanner: Lizard Morphology
45
muscle pass posteromedial, whereas in the other three genera they pass transversely or fan out.
The intemnandihukiris anterior profundus (Figs. 5, 6) occupies a position posterior and dorsal to the intermandibularis anterior su- perficialis, and anterior and dorsal to the in- termandibularis posterior. In Uma the fibers extend medially, with only a few fibers at each end fanning out. In Cophosaurus it is distinctly fan shaped, and in Callisaurus it fans out to a lesser degree. In Holbrookia it is slightly fan shaped, with most fibers slant- ing posteriorly. There were two areas of variation seen in the neck musculature.
The constrictor colli (Figs. 5, 6) is the su- perficial muscle of the neck, originating in the dorsolateral fascia of the neck and in- serting in the ventral raphe of the throat posterior to the intermandibularis posterior. It is one or two muscle fibers thick, rather narrow, and the origin is broader than the insertion.
In all four genera the muscle fibers of the constrictor colli reach the mid-throat area, but they originate in a facia without reach- ing the dorsal skeletoginous septum.
The constrictor colli is widest in Uma. It covers from the posterior edge of the tym- panum to the shoulder, nearly covering the depressor mandibularis completely. In the other three genera it is more straplike and only fills half the space between the posteri- or edge of the tympanum and shoulder, being centered in this area.
The depressor mandibularis (Figs. 5, 6) originates in the mid-dorsal raphe, along the posterior borders of the parietal bone and parietal wings. The insertion has three slips; one inserts deep to the pterygomandibularis on the articular process of the lower jaw. Another passes superficially to the pterygo- mandibularis and the intermandibularis pos- terior; it inserts on the ventrolateral surface of the mandibular rami by interdigitating at right angles with these muscles. The third slip inserts on both sides of a tendon that extends dorsally from the articular process. The attachment of this slip to the articular is superficial to the deep slip mentioned above. In the earless lizards (Cophosaurus and Holbrookia) a portion of this muscle is
expanded anteriorly to partially cover the tympanic cavity.
The sternohyoideus I and omohyoideus are closely related, and there is considerable confusion in the literature concerning them. The position taken here is that those por- tions that originate on the scapula and cla- vicle are omohyoideus, and those portions that originate on the sternum are ster- nohyoideus.
Both muscles are deep to the constrictor colli, and the episterno-cleidomastoideus. The muscle dorsal to them is the sterno- hyoideus II.
The omohyoideus (Fig. 6) takes two forms; in Uma and most species of Hol- brookia it has a single head originating on the scapula and clavicle. In Callisaurus, Cophosaurus, and some Holbrookia the omo- hyoideus is divided for its entire length, connected only by a myocomma located midway between the origin and insertion, bisecting the muscle and binding the muscle fibers together. This myocomma also bisects the sternohyoideus I, and the two muscles are bound firmly together by it.
The omohyoideus can be easily separated from the sternohyoideus posterior to the myocomma, but they cannot be distin- guished anterior to it except by position.
The sternohyoideus I (Fig. 6) originates on the sternum and inserts on the proximal end of the ceratobranchial I and II. The myocomma is at its junction with the omo- hyoideus, and it unites the two muscles. The medial portion of the omohyoideus cannot be distinguished from the sternohyoideus an- terior to this myocomma in all genera.
Statistical Analysis
The osteology of the skulls was analyzed by taking 43 measurements from each skull and by calculating all possible ratios. The Statistical Package for the Social Sciences (SPSS) discriminant analysis was employed to determine which ratios were of value in distinguishing genera and if the lizards could be classified by using them. The theo- ry and use of discriminant analysis is de- scribed by Klecka (1975). From each skull 903 ratios were generated, and those suf-
46
Great Basin Naturalist
Vol. 37, No.
ficient to separate one genus from the other three were used.
The discriminant analysis proceeds in a stepwise fashion by selecting the single best- discriminating variable, and then selects a second on the basis of its ability to improve the value of the discrimination criterion in combination with the first variable. Third and subsequent variables are similarly se- lected according to their ability to contrib- ute to further discrimination. A plot of the discriminant score in two dimensions, a ter- ritorial map, and a classification based on the preceding analysis are given. Range, mean, and standard deviation of ratios for each genus are also given, and examples are presented in Figs. 7 and 8 along with a plot of the ratios (Figs. 7-11) that illustrates rel- ative relationships between genera. These figures are only examples of the ratios, and were selected because they clearly demon- strate the results referred to below.
A multivariate analysis of variance was performed to determine the significance of differences between the four genera. The comparisons were based on a non-orthogan- al set such that the differences between Uma and the other three genera were test- ed (comparison #1), the differences be- tween CaUisaurus and Cophosaums were tested (comparison #2), and, finally, the differences between Cophosaums and Hol- brookia were tested (comparison #3).
Using 116 ratios selected in the discrimi-
nant analysis, it is possible to classify the four genera. Of 21 tests, classification agreed with the present taxonomic system in 17 of them. The first classification error was in Test 7, where a CaUisaurus and a Cophosaums were both classified as Hol- brookia. Only three ratios were involved, and all three involved the squamosal width divided by the vomer width, palatine length, and palatine width respectively. These ratios were effective in separating Uma from the other three, but were not useful in distinguishing between the other three genera.
The second classification error occurred in Test 8. Here a CaUisaurus and a Hol- hrookia were both classified as Cophosaums. Four ratios were involved, and each one used the squamosal width divided by the pterygoid width, the epipterygoid width, the hyoid length, and the hyoid width. Whereas Uma was well separated in Test 7, it was not well separated in Test 8.
The most confused classification was in Test 15. Here one CaUisaurus and one Hol- brookia were both classified as Cophosaums, one CaUisaurus was classified as Uma, one Cophosaurus was classified as Holbrookia, and one Holbrookia was classified as CaUi- saurus. Two ratios were used, and they were derived from the mandible width di- vided by the frontal width and the nasal width. Uma is the most distinct in this test; however, the individual plot scores show
|
CalUsa |
mrus |
9.6875 |
to 12.3636 |
10.8422 |
1.1974 |
Callis |
aurus |
4.5405 to 5.6667 |
5.0379 |
0.4923 |
|
Cophosa |
lurus |
10.3333 |
to 13.8869 |
11.7222 |
1.5437 |
Cophos |
aurus |
5.2963 to 5.6000 |
5.4190 |
0.1337 |
|
Holbroc |
Jcia |
11.5000 5.4375 |
to 13.7500 to 7.1379 |
12.1583 6.1048 |
1.1751 0.7827 |
Holbro |
okia |
4.7500 to 6.7647 4.0200 to 4.3500 |
5.5560 4.1624 |
0.9255 |
|
^ |
0.1373 |
|||||||||
|
Total |
5.4375 |
to 13.8889 |
10.2069 |
2.7175 |
Total |
4.0200 to 6.7647 |
5.0438 |
0.7361 |
Fig. 7. Ratio of the skull length /squamosal width used in Test #5 of the Discriminant Analysis. Range, mean, and standard deviation plotted for the ratio of skull length/squamosal width in the four current gen- era.
Fig. 8. Ratio of the skull length /mandible width used in Test # 14 of the Discriminant Analysis. Range, mean, and standard deviation plotted for the ratio of skull length /mandible width in the four current gen-
March 1977
Cox, Tanner: Lizard Morphology
47
Fig. 9. Range, mean, and standard deviation in the four current genera plotted for the ratios of: A. squa- mosal length /squamosal width; B. squamosal width /quadrate length; C. jugal length/squamosal width; D. squamosal width/mandible length.
considerable variation, and there is no dis- crete grouping to distinguish the separate genera.
The last confused classification was in Test 17. Here four ratios were used that in- volved the mandible width divided by the palatine length, palatine width, epiptery- goid length, and the mandible length. In this classification a Callisaurus and Co- phosaurus were both mistaken for Hol- brookia. The plot indicates a complete sepa- ration of Uma, but with an overlap of the other three genera.
Over all 21 tests the centroids were sepa- rated on the average of the following dis- tances: Uma was separated from Callisaurus by 2.10 mm, from Cophosaurus by 2.44 mm, and from Holbrookia by 2.67 mm; Cal- lisaurus was separated from Cophosaurus by 1.19 mm and from Holbrookia by 1.58 mm; and Cophosaurus was separated from Hol- brookia by 0.92 mm. These centroids are derived from the discriminant functions, which in turn are derived from linear com- binations of the variables used in each test. The purpose for the discriminant analysis was to find ratios that would discriminate
Holbrookia
Fig. 10. Range, mean, and standard deviation in the four current genera plotted for the ratios of: A. quad- rate length /quadrate width; B. maxillary length /quadrate length; C. jugal length /quadrate length; D. skull length /quadrate length.
Fig. 11. Range, mean, and standard deviation in the four current genera plotted for the ratios of: A. man- dible length/mandible width; B. jugal length /mandible length; C. skull length/jugal length.
48
Great Basin Naturalist
Vol. 37, No. 1
between the groups; this goal was achieved. It is also desirable to determine how well each group is defined, and if it is signifi- cantly different from all other groups. The SPSS program does not provide this infor- mation, but it is possible, by studying the plots (Figs. 7-11), to see that the range of Callisaurus overlaps the range of Holhrookia and/or Cophosaurus in nearly all ratios plotted. The range for Uma is only occa- sionally overlapped by the other genera, and it is usually only Callisaurus that over- laps Uma.
The multivariate analysis of variance was performed to determine the degree of sig- nificance between the four genera. The non- orthoganal set of comparisons was as fol- lows:
Comp. Cal Coph.
Hoi.
Uma
1 1
2 -1
3 0
1 1 -3
1 0 0
1 -1 0
The degrees of freedom, F-ratios, and actual probability for each F-value are presented in Table 1.
The differences between Uma and the other three genera are highly significant, but the differences between the remaining three genera are only slightly significant, and in many cases the differences are in- significant.
Uma is therefore the best differentiated by these tests. The other three genera are close together, and although they can be distinguished on the computer, their degree of separation is slight.
The probability of error in differentiating between Callisaurus and Cophosaurus is similar to the probability of error in differ- entiating between Cophosaurus and Hol- hrookia (average between 9 percent and 12 percent), verifying these statements.
These data are based on 116 ratios that were found to be the best discriminators of the 903 ratios examined. It is noteworthy that by using the very best osteological dis- criminators, the genera are often not clearly separated.
D
ISCUSSION
Sand lizards are a closely related group according to Smith (1946), Norris (1958), Axtell (1958), Etheridge (1964), and Clarke (1965). Smith referred to them as a closely knit group and listed their common charac- teristics as being oblique labials, granular dorsal scales, small head scales, a gular fold, a peculiar median triangular postmental, several prominent postlabials, much the same habits, similar habitat, scoop-shaped heads, flaring labial regions, and a counter- sunk lower jaw. Axtell (1958) listed 22 char- acters that he felt would describe ancestral sand lizards. With only a few exceptions, this list of characters might just as well ap- ply to other sceloporine lizards.
Callisaurus, Cophosaurus, and Holhrookia exhibit many of the primitive characteristics mentioned by Axtell. Each genus, however, shows specializations that vary from those listed. Callisaurus is similar to Axtell's cri-
Table 1. Summary of F-ratios from eight multivariate analyses of variance tests.
|
Comparisons |
|||
|
1 |
2 |
3 |
|
|
Test df. |
F-ratio/Probability |
F-ratio/Probability |
F-ratio/Probability |
|
1 6/7 |
7.797/0.79% |
1.858/21.8% |
4.034/4.51% |
|
2 7/6 |
11.055/0.468% |
4.623/4.05% |
1.973/21.29% |
|
3 7/6 |
25.922/0.044% |
4.596/4.106% |
31.230/0.026% |
|
4 5/8 |
96.377/0.000062% |
1.141/41.24% |
2.827/9.25% |
|
5 6/7 |
12..387/0.20% |
16.8.30/0.077% |
4.450/3.56% |
|
6 5/8 |
40.174/0.00185% |
2.100/16.76% |
6.117/2.54% |
|
7 6/7 |
10.251 /0.,35% |
14.480/0.124% |
2.147/17.02% |
|
8 4/9 |
7.807/0.532% |
4.670/2.572% |
2.118/16.07% |
March 1977
Cox, Tanner: Lizard Morphology
49
teria for the primitive condition, with the most striking speciahzations being increased length of tail and limbs and a more slender body form. Holbrookia is also similar to an- cestral sand lizards except for the covered tympanum. Body form in Holbrookia is not as proportionately slender as it is in Calli- saurus, nor is the tail as long. The problem- atic Cophosaums has characteristics of both. Its body form, limbs, and tail approach those of Callisaurus, but it has a covered tympanum like Holbrookia. Uma has devel- oped, to a greater degree, the dorsovent- rally flattened body and the toe fringes. Callisaunis, Holbrookia, and Cophosaurus occupy generalized habitats, probably sim- ilar to that of the ancestral stock. Uma, on the other hand, is restricted to a sand dune habitat.
Osteology
Comparisons of skulls of Ctenosaura (Oelrich 1956), as well as general accounts of reptile osteology by Williston (1925), Ro- mer (1956), and Avery and Tanner (1964, 1971) indicated that osteological character- istics of iguanid lizard skulls are generally stable within a genus. Studies on Sauro- malus (Avery and Tanner 1964), Croto- phytus (Robison and Tanner 1962), and Ctenosaura (Oelrich 1956) portrayed the ap- parent general stability of osteological char- acters found in iguanid skulls; however, skulls of Callisaurus, Cophosaurus, Hol- brookia, and Uma observed in this study are peculiar to iguanid skulls as portrayed by the above authors because there is stability within the sand lizard genera as a group rather than within a single genus. The lac- rimals and postfrontals are also absent in all the sand lizards, a condition not generally found in iguanid skulls, except that Jenkins and Tanner (1968) found that two species groups of phrynosoma also lack these skull bones. Etheridge (1964) pointed out the ab- sence of the lacrimals and postfrontals in sand lizards.
Deviation by sand lizards from the gener- al iguanid skull, particularly evidenced by the fusion or loss of the lacrimal and post- frontal bones, is evidence supporting the hy-
pothesis that sand lizards are highly special- ized among the iguanids.
In discussing the osteology of Holbrookia, Axtell (1958:24) stated:
In general the osteology of the genera Calli- saurus and Vma corresponds closely with that of Holbrookia, so this discourse may apply just as well to the entire sand lizard section of the family iguanidae.
Development of the covered tympanum appears to be related to the osteological variations observed in this study. The quad- rate, squamosal, paraoccipital process, and mandible are all adjacent to the ear, and all exhibit variations (Figs. 1, 2, .3, and 4). With the loss of the external ear, the quadrate is reduced in size, and the paraoccipital proc- ess is directed forward, the mandible is shortened, the squamosal is narrowed, and the hyoid is also shortened. These modifica- tions of the skull are apparent in earless sand lizards and are less modified in sand lizards having an external ear. Callisaurus and Uma are alike in that the paraoccipital process is directed caudad, the quadrate proportionately larger, and the mandible long and wide in comparison to the earless forms (Figs. 1, 2, 3, and 4). In Uma the de- gree of development of these characters is different because the squamosal is wide. A ratio of skull length divided by squamosal width shows: (a) Uma ranging from 5.4375 to 7.1379; (b) Callisaurus ranging from 9.6975 to 12.3636; (c) Cophosaurus ranging from 10.3333 to 13.8889; and (d) Holbrookia ranging from 11.0833 to 13.7500 (from Fig. 7). The quadrate and mandible length and width are similarly enlarged in Uma when compared to the other groups. These differ- ences are sufficient to permit the computer to distinguish between these measurements in Uma and those same characteristics found in Callisaurus and the earless group.
There is some variation in the position and articulation between the squamosal, jugal, and postorbital. In all except some Uma individuals, the jugal reaches the squa- mosal. In Callisaurus it edges between the squamosal and postorbital. In Holbrookia and Cophosaurus the degree of encroach- ment is increased. This may be a result of the forward direction taken by the para-
50
Great Basin Naturalist
Vol. 37, No. 1
occipital process and the shortening of the jaw. Uma is unique in that the jugal fails to reach the squamosal in some individuals, in which case the postorbital is found wedg- ing between them (Fig. 3). In other Uma individuals a mere contact is made without any overlap of the squamosal and jugal.
That these variations exist is verified by the statistical analysis. The SPSS discrimi- nant analysis was able to identify character- istics that were capable of separating the genera into distinct groups. However, the analysis of variance points out that differen- ces between Callisaurtis, Copho.saurus, and Holhrookia (as identified by SPSS) are insig- nificant and that Ihna is indeed a distinct group. The F-ratios derived from the analy- sis of variance illustrate these facts. The tests comparing Uma with the other three genera have high F-values (96.377 to 7.797) and consequently low probability of making classification errors (from 0.000062 percent to 0.79 percent). Therefore the differences between Uma and the other sand lizards are highly significant. In comparing Callisaurus with Cophosaurus, the F-values were found to be lower (16.830 to 1.141), and con- sequently the probability of classification er- ror is higher (41.24 percent to 0.077 per- cent). Only two of the eight tests were significant (below 2.5 percent probability of error). The tests comparing Cophosaurus with Holhrookia are similar to those com- paring Callisaurus and Cophosaurus. The F- ratios range from 31.231 to 1.973. The probability of error is again higher (21.29 percent to 0.026 percent). Low F-ratios and high error probability indicate that the dif- ferences between the three genera are in- significant.
Myology
The literature is void of studies dealing directly with myology of sand lizards. Earle (1961a, 1961b, 1961c, 1962) dealt with the middle ear and also touched on the myol- ogy in the ear region. In this study, we will deal only with the myology of the head and neck region. A comparison of the anterior anatomy of sand lizards shows some obser-
vable differences that are discussed below (Figs. 5 and 6).
The intermandibularis muscles show some variations that are useful in distinguishing the genera. Uma is most distinct, with its intermandibularis anterior superficialis mus- cle being fan shaped (Fig. 6). This condition is contrasted with that found in Holhrookia, where the transverse mandibulae muscles are so nearly parallel that the borders of each muscle are difficult to discern. Co- phosaurus and Callisaurus exhibit an inter- mediate condition with some fanning seen, but not to the extent seen in Uma. Co- phosaurus can be distinguished from Calli- saurus in that fibers of the intermandicularis anterior superficialis extend posteromedially, and in Cophosaurus, a portion of the in- sertion of the intermandibularis anterior profundus is found anterior to the super- ficialis, a condition not seen in Callisaurus.
The depressor mandibulae also shows some variations. In the earless lizards there is an expanded anterior edge of this muscle, which partially covers the enclosed tympa- num. This muscle emerges from beneath the constrictor colli to insert upon the man- dible, with its most ventral fibers inter- digitating at right angles with the inter- mandibularis posterior. Furthermore, the depressor mandibulae fibers extend much further along the mandibular ramus than in the eared genera. In Cophosaurus these fi- bers reach the mandibular ramus with only a few interdigitating with the inter- mandibularis posterior.
The omohyoideus is also variable in these genera. In the unfolding of the evolutionary development of the omohyoideus, a branch of the rectus cervicus originated on the sca- pula and inserted on the hyoid. In sand liz- ards it further developed into a complex of muscles. In Uma the omohyoideus has a second head that originates on the sternum. This then can be called sternohyoideus I. Another muscle, the sternohyoideus II, arises on the sternum, passes deep to the sternohyoideus I, and inserts on the pos- terodorsal edge of the ceratobranchial I. Avery and Tanner (1964) designated this muscle thyrohyoideus in Sauromalus, but, since it clearly originated on the sternum in
March 1977
Cox, Tanner: Lizard Morphology
51
sand lizards, it is designated sternohyoideus II in this study.
A division of the oniohyoidens has oc- curred in CaUisaurus, Cophosaiirus, and two specimens of Holbrookia {Holbrookia lacerata CAS 73979 and Holbrookia m. approximans BYU 17099). Thus, in sand lizards the muscle may have one to three origins: the most lateral dorsal one is on the scapula, the second is on the clavicle, and the most me- dial is on the sternum and interclavicle. The omohyoideus also has two insertions: the most lateral division inserts on the distal two-thirds of the ceratobranchial I, and the second and medial heads unite and insert on the proximal third of the ceratobranchial II and the basihyal. In Uma and for the most part in Holbrookia, such division of the omohyoideus has not occurred, and the in- sertion is continuous along the ceratobran- chial I, the basihyal, and the proximal third of the ceratobranchial II. These myological variations, although observable, are not of great magnitude and may not be sufficient to support generic status for CaUisaurus, Cophosaurus, or Holbrookia. In all there is an overwhelming myological similarity within these three genera, and they are ob- viously closely related to Uma.
Phylogeny
Norris (1958) and Axtell (1958) agreed that sand lizards began their radiation in early or middle Miocene. This is the time when the Sierra Madre Occidental Range of Mexico was being built by volcanism along the western and southern borders of the Mesa of Central Mexico (Schuchert 1935, Miller 1942). Axtell indicated that this vol- canism split the sand lizards into two groups, a Uma-Callisaurus group and a Hol- brookia prototype. Norris also believed the sand lizards were divided at this time, but indicated that Uma was isolated from the CaUisaurus- Holbrookia stock. Norris further indicated that Uma was subsequently split by continued mountain-building processes in the mid-Pliocene, giving rise to the exsul group and notata-scoparia stocks. He also postulated that it was during this same time
that the CaUisaurus- Holbrookia stock was divided. The data available to us from this study tend to support Norris' concepts con- cerning the radiation of sand lizards. Ac- cording to Maslin (1952), "The basic as- sumption upon which all taxonomic practices rest is that similar organisms are related." If we assume that sand lizards have occupied similar habitats and have been subject to similar environments throughout their recent history, then we can assume that the degree of similarity be- tween groups is an indication of the close- ness of their relationship. Maslin (1952) also states that because internal characters are less variable, they are of much greater val- ue in establishing relationships than are ex- ternal characters. The internal characters considered here indicate a closer relation- ship between Holbrookia and CaUisaurus than between Uma and CaUisaurus, in- dicating that Uma was probably the first of the sand lizards to break away from the an- cestral line.
Clarke (1965) examined the behavior and external morphology from the standpoint of 20 characters; in 14 of them Cophosaurus was like CaUisaurus, in 5 of them Co- phosaurus was unique, and in only one was Cophosaurus like Holbrookia. In discussing these comparisons he states:
The distinctness of Cophosaurus is evident. It is intermediate in many features between CaUi- saurus and Holbrookia, with the data indicating a closer affinity to CaUisaurus than to Hol- brookia. The uniqueness of Cophosaurus is most clearly shown in the distinctness of the push-up pattern.
On the basis of five characters, Clarke would separate Cophosaurus as a separate genus. These five characters are: (1) place- ment of the lateral bar: anterior for CaUi- saurus, posterior for Cophosaurus, and cen- tral for Holbrookia; (2) body shape: slender for CaUisaurus, intermediate for Co- phosaurus, and stout for Holbrookia; (3) middle ear: distinct with an external open- ing for CaUisaurus, distinct without an ex- ternal opening for Holbrookia, and inter- mediate without an external opening for Cophosaurus; (4) preferred body temper- ature: 39.2 C for CaUisaurus, .38.3 C for
52
Great Basin Naturalist
Vol. 37, No. 1
Cophosaurtis, and 35.7-38.1 C for Hol- hrookia; and (5) push-up pattern: distinct in all three groups. In four of the five charac- ters the differences are only comparative, and do not indicate a clear-cut distinction; only in the fifth does Cophosauriis show a real distinctness.
An alternate interpretation of these data would have to conclude that there is not enough difference to warrant generic status for Cophosaurtis; indeed, there also may not be enough difference to warrant generic status for Holbrookia. The variations de- scribed are of the kind and magnitude used in the descriptions of species. Clarke (1965) stated that the push-up pattern is the most distinct feature of Cophosaurtis. Carpenter (1963, 1967) described the same behavior for Uma, indicating that a genus is capable of supporting greater variation than Clarke has allowed for in Callisaurus, Cophosaurus, or Holbrookia.
Guttman (1970b) also commented on Clarke's study, stating:
A comparison of Callisaurus, Cophosaurus, and Holbrookia (Clarke 1965) indicated the great similarity among these genera. According to Clarke, the uniqueness of Cophosaurus is most clearly shown by its distinctive push-up pattern. A comparison of the display-action pat- terns of two species of Urosaurus (Carpenter 1962) or three species of Uma (Carpenter 1963) indicated to this writer that sufficient in- trageneric variation exists to refrain from estab- lishing a genus based on this display pattern.
Guttman was reporting his electrophoretic study of sand lizards, in which he analyzed the hemoglobin components and found that they were all identical. This is highly un- usual, especially for different genera. Elec- trophoretic techniques have been of value in confirming taxonomic relationships. This has been demonstrated by Dessaur et al. (1962), Dcssauer (1966), Gorman and Des- sauer (1965), Gorman (1966), Maldonado and Ortez (1966), and Guttman (1970a and 1970b). The conclusion Guttman (1970b) came to, and the one that supports our con- clusions, was that the .sand lizards are more closely related than their present taxonomic status indicates.
Simpson (1945) states:
What is deplorable in splitting is the tenden- cy to raise the ranks of groups without need, that is, without gaining any practical advan- tage. One of the more evident symptoms of this tendency is the appearance of many monotypic groups in classification.
The proposal resulting from this study and data examined from other studies would eliminate one monotypic genus and would provide a better indication of the close rela- tionships that are so evident in the sand liz- ards.
Axtell (1958) believed that the sand liz- ards evolved under subhumid conditions, not greatly different than the conditions ex- isting today in the sand lizard range. He postulates that it was during the mid-Plio- cene that Holbrookia developed the covered tympanum. He then states, "The species previously known as Holbrookia texana, but which now appears to belong in the Calli- saurus line of evolution, has probably devel- oped the covered tympanum independ- ently." Axtell's phylogenetic tree for the sand lizards is presented in Figure 12.
Earl (1961a, 1961b, 1961c, and 1962) in- dicated that Cophosaurus was intermediate between Callisaurus and Holbrookia in ear anatomy, but agreed with Axtell that ear- lessness evolved twice, reporting that re-
^ DIRECTION OF SPECIALIZATION 3,
TYMPANUM COVERED
SAND LIZARD PROGENITOR
Fig. 12. Phylogeny of the sand lizards according to Axtell (1958).'
March 1977
Cox, Tanner: Lizard Morphology
53
lated groups have the potentiaHty to devel- op identical clines, and under similar environmental conditions these identical clines may develop at different times and places. This concept was discussed by Mas- lin (1952), who put forth the idea while dis- cussing morphological criteria of phyletic relationships. Norris (1958) and Clarke (1965) also agreed with Axtell about the idea of separate earless evolution.
The concept of two evolutions for earless- ness may be a major barrier to the under- standing of sand lizard relationships. Since we lack a fossil record of sand lizards, there is no way of knowing when or how such a character came about. It has been suggested (Earl 1961a) that it came about in response to the burrowing habit, but this is purely speculative, as there are many burrowing species that do not have a covered tympa- num (including Uma). A covered tympanum is not unique to "sand lizards " because the agamid genus Tympanocryptis in Australia is earless, and so are some of the members of Phrynocephahis. Phrynosoma has both eared and earless members. Norris's (1958) idea is acceptable when he says that CalU- saurus and Holbrookia split in the early Pliocene, with Holbrookia occupying the table lands of the mesa of Central Mexico, and Callisaurus having been isolated from Holbrookia before their radiation to the more northern habitats. The habitats of the Chihuahuan desert and Sonoran desert are similar; the draconoides and texanus groups would have had an excellent opportunity to parallel each other sufficiently to account for the external morphologic similarities. In- ternal structures are not as accessible to ex- ternal selective pressures and may, then, in- dicate more accurately the true relationships: that the texanus group is more closely related to Holbrookia than to Callisaurus, which it resembles through par- allelism.
That earlessness may have evolved twice is possible, and it is most probable that there were separate evolutions for this char- acter in the genera Tympanocryptis, Phry- nocephalus, Phrynosoma, and sand lizards. To theorize that it evolved twice in the sand lizards is, however, a questionable con-
cept. Earless sand lizards are closely related, their geographic ranges overlap, their habit- ats are similar, their food requirements are similar, the ear anatomy is similar, and we lack any evidence from fossil records that they diverged before the earless character arose. Where evidence indicates a close tax- onomic relationship, as it does here, the idea of a single evolution for the earless character is most plausible. It is very un- likely that the same character would evolve twice in the same way in two groups that are as closely related as are earless sand liz- ards. (A proposed phylogenetic tree is given in Figure 13.)
These data, when added to that of earlier workers, seem to clearly indicate that sand lizards may best be represented by three genera: Uma, Callisaurus (as at present con- stituted), and Holbrookia.
The characteristics that separate sand liz- ards into genera are few and not well de- fined when compared to distinctions be- tween other Sceloporine genera. The relationship between Callisaurus and Hol- brookia is especially close, with earlessness (and its associated skull characters) and body proportions being the most striking variants. Uma, on the other hand, appears to be well defined. It is felt that Holbrookia is, therefore, a recent derivative of Calli-
^ DIRECTION OF SPECIAIIZATION ^
TYMPANUM COVERED
SAND LIZARD PROGENITOR
Fig. 13. Proposed phylogeny of the sand lizards as dictated by data developed from this study.
54
Great Basin Naturalist
Vol. 37, No. 1
saiirus, evolving earlessness and an adapta- tion for varied habitats in a relatively short time, perhaps since their separation in the late Pliocene.
Sand lizards, particularly the species and subspecies, particularly in H. maculata, in- dicate that this group has recently under- gone adaptive radiation. If recent geological past has been correctly interpreted by re- cent paleontological findings (Etheridge 1961, Wells and Jorgensen 1964), the desert areas of today, extending from Texas to California, were very different as recently as 10,000 years ago, indicating that sand liz- ard adaptative radiation must be relatively recent.
Evidence from internal morphology and geographical distribution indicates that ear- less sand lizards should remain as two close- ly related groups in the genus Holbrookia. Data from comparative skull and throat anatomy, if used alone, indicate a very close relationship between all genera in the sand lizard group, but it is felt that the dis- tinctions, however small, do indicate that Holbrookia has evolved from CaUisaums stock and has achieved sufficient distinctness to be given generic status. It is, therefore, proposed that sand lizards be classified as they were before Cophosaiirus was split off by Clarke (1965). This is as follows:
Uma notata Baird
Uma scoparia Cope
Uma exsul Schmidt
Uma pamphijgas Williams, Chrapliny, and Smith
CaUisaums draconoides Blainville
Holbrookia texana Troschell
Holbrookia lacerata Cope
Holbrookia maculata Girard
Holbrookia propinqiia Baird and Girard
Literature Cited
Avery, D. F., and W. W. Ta.nner. 1964. The os- teology and myology of the head and thorax re- gions of the obesus group of the genus Sauro- maliis Dumeril (Iguanidae). Brigham Young Univ. Sci. Bull., Biol. Ser. 5(.3):l-30.
1971. Evolution of the iguanine lizards (Sauria
Iguanidae) as determined by osteological and myological characters. BYU Sci. Bull., Biol. Ser. 12(3): 1-79.
AxTELL, B. VV. 1956. A .solution to the long neglected Holbrookia lacerata problem and the description
of two new subspecies of Holbrookia. Bull. Chi- cago Acad. Sci. 10(11):161-179.
1958. A monographic revision of the iguanid
genus Holbrookia. Diss. Abstr. 19(6): 1476-1477.
1960. Orientation by Holbrookia maculata
(Lacertilia, Iguanidae) to solar and reflected heat. Southwestern Naturalist 5(l):47-48.
Baird, S. F. 1859. Description of new genera and spe- cies of North American lizards in the museum of the Smithsonian Institution (Uma). Proc. Acad. Nat. Sci. Philadelphia 10:253 (1858).
Baird, S. F., and C. Girard. 1852. Holbrookia texana. Proc. Acad. Nat. Sci. Philadelphia 6:124.
Barbour, T. 1921. A new lizard from Guaymas, Mexi- co. Proc. New England Zool. Club 7:79-80.
Blainville, H. M. D. de. 1835. Description de quelques especes de reptiles de la Californie precedee de I'analyse dun systeme general d'herpetologie et d'amphibiologie. Nouv. Ann. Mus. Natn. Hist. Nat. Paris, (3)4:232-296.
BocouRT, M. F. 1873-1897. Etudes sur les reptiles. Mission scientifique au Mexique et dans FAmerique Central-Becherches Zoologiques, (q.v.) part 3.
Bogert, C. M., and E. E. Dorsom. 1942. .\ new lizard of the genus Callisaurus from Sonora. Copeia 1942(3): 17,3-175.
Burt, C. E. 1931a. On the occurrence of a throat-fan in the sand lizard, Uma notata Baird, with notes on the adaptive specialization of the form. Copeia 1931(1): 1.5-16.
1931b. On the occurrence of a throat-fan in
Callisaurus ventralis gabbii and two species of Crotaplnjtus. Copeia 1931(2):58.
Cagle, F. R. 1950. Notes on Holbrookia texana in Texas. Copeia 1931(2):58.
Carpenter, C. C. 1962. A comparison of the patterns of display of Urosaurus, Uta, and Streptosaurus. Herpetologica 18:145-152.
1963. Patterns of behavior in three forms of
the fringe-toed lizards, Uma. Copeia 1963(2):406-412.
1967. Display patterns of the Mexican iguanid
lizards of the genus Uma. Herpetologica 23(4): 285-293.
Clarke, R. F. 1965. An ethological study of the igua- nid lizard genus Callisaurus, Cophosaurus, and Holbrookia. The Emporia State Research Studies 13(4): 1-66.
Cope, E. D. 1880. On the zoological position of Texas. Bull. U.S. Nat. Mus., No. 17:1-51.
1883. Notes on the geograhical distribution of
Batrachia and Reptilia in western North Amer- ica. Proc. Acad. Nat. Sci. Philadelphia 36:10-11.
1894. On the iguanian genus Uma Baird. Am.
Nat. 28:4.34-435.
1895. On the species of Uma and Xantusia.
Am. Nat. 29:938-939.
1896. On the genus Callisaurus. Am. Nat.
.30:1049-1050.
1900. The crocodilians, lizards, and snakes of
North America. Report U.S. Nat. Mus. for 1898, pp. 15.3-1270.
March 1977
Cox, Tanner: Lizard Morphology
55
Dessauer, H. C. 1966. Taxonomic significance of elec- trophoretic patterns of animal sera. Rutgers Univ., Serai. Mus. Bull. 34:4-8.
Dessauer, H. C, W. Fox, and F. H. Pauch. 1962. Starch-gel electrophoresis of transferrins, este- rases, and other plasma proteins of hybrids be- tween two subspecies of whiptail lizard (genus Cnemidaphoms). Copeia 1962:767-775.
DicKERsoN, M. C. 1919. Diagnoses of twentv-three new species and a new genus of Lizards from lower California. Bull. Mus. Nat. Hist., 41(10):461-477.
Earle, a. M. 1961a. The middle ear of Holbrookia maculata maciilata, the northern earless lizard. Copeia No. 1:68-74.
1961b. An additional note on the ear of Hol- brookia maculata. Copeia No. .3:355.
1961c. The middle ear of Holbrookia and Cal-
lisaitrus. Copeia 1961(4):405-410.
1962. The middle ear of the genus Uma com- pared to those of other sand lizards. Copeia 1962(1): 185-188.
Etheridge, R. 1961. Late Cenozoic glass lizards (Oph- isaurus) from the southern Great Plains. Herpe- tologica. 17(3): 179-186.
1964. The skeletal morphology and systematic
relationships of sceloporine lizards. Copeia 1964(4):610.
Girard, C. F. 1851. On a new American saurian rep- tile. Proc. Am. Assoc. Advmt. Sci., 4:200-202.
Gorman, G. C. 1966. The relationships of Anolis of the Roquet species group (Sauria: Iguanidae), electrophoretic comparison of blood proteins. Comp. Biochem. Physiol. 19:845-853.
Gorman, G. C, and H. C. Dessauer. 1965. Hemoglo- bin and transferrin electrophoresis and relation- ships of island populations of Anolis lizards. Sci- ence 150:1454-1455.
GuTTMAN, S. I. 1970a. Hemoglobin electrophoresis and relationships within the lizard genus Sceloporus (Sauria: Iguanidae). Comp. Biochem. Physiol. 34:563-568.
1970b. An electrophoretic .study of the he- moglobins of the sand lizards, Callisaurus, Co- phosaurus, Holbrookia, and Uma. Comp. Bio- chem. Physiol. 34(3):569-574.
Harper, F. 1932. A new Texas subspecies of the lizard genus Holbrookia. Proc. Biol. Soc. Washington (45): 15-18.
Heifetz, W. 1941. A review of the lizards of the genus Ihna. Copeia 1941 (2):99-lll.
Jenkins, R. L., and W. W. Tanner. 1968. Osteology and myology of Phrynosoma P. Platijrhinos Gi- rard and Phrynosoma d. hernondesi Girard. BYU Sci. Bull., Biol. Ser. 9(4): 1-34.
JuDD, F. W. 1974. Intraspecific variation in blood properties of the keeled earless lizard, Hol- brookia propinqua. Herpetologica .30(1):99-102.
Klecka, W. R. 1975. Discriminant analysis, pp. 434- 467. In. N. H. Nie, ed.. Statistical Package for the Social Sciences, second edition. McGraw- Hill Book Company, New York.
Lannon, J. R. 1962. A different method of catching
the desert lizards, Callisaurus and Vma. Copeia 1962(2):437-4.38.
Larsen, K. R., and W. W. Tanner. 1974. Numeric analysis of the lizard genus Sceloparus with spe- cial reference to cranial osteology. Great Basin Nat. 34(1):1-41.
1975. Evolution of the sceloporine lizards
(Iguanidae). Great Basin Nat. .3.5(1): 1-20.
Linsdale, J. M. 1940. Amphibians and reptiles in Ne- vada. Proc. Am. Acad. Arts Sci. 73(8): 197-257.
Maldonado, a., and E. Ortiz. 1966. Electrophoretic patterns of serum proteins of some West Indian Anolis (Sauria: Iguanidae). Copeia 1966(2): 179- 182.
Maslin, p. T. 1952. Morphological criteria of phyletic relatioiLships. Systematic Zoology 1:49-70.
Mayhew, W. W. 1964. Photoperiodic responses in three species of the lizard Uma. Herpetologica 20(2):95-113.
1964. Taxonomic status of California popu- lations of the lizard genus Uma. Herpetologica 20(3):170-ia3.
1966. Reproduction in the arenicolous lizard
Uma notata. Ecology 47(1): 9- 18.
NoRRis, K. S. 1958. The evolution and systematics of the iguanid genus Uma and its relation to the evolution of other North American desert rep- tiles. Bull. Am. Mus. Nat. Hist. 114(3):247-326, Figs. 1-17.
1975. Activity and thermal ecology of the
keeled earless lizard, Holbrookia propinqua. Herpetologica 31(2): 137- 149.
Oelrich, T. M. 1956. The anatomy of the head of Ctenosaura pectinata (Iguanidae). Misc. Pub. Mus. Zool. Univ. Michigan 94:1-122, 59 figs.
Peters, J. A. 1951. Studies on the lizard Holbrookia texana (Troschel) with descriptions of two new subspecies. Occ. Papers Mus. Zool. Univ. Mich- igan. No. 537:1-20.
PiANKA, E. R., AND W. S. Parker. 1972. Ecology of the iguanid lizard Callisaurus draconoides. Copeia 1972:493-508.
Presch, W. 1969. Evolutionary osteology and relation- ships of the horned lizard genus Phrynosoma (Family Iguanidae). Copeia 1969(2):250-"275.
Ramsey, L. W. 1948. Hibernation of Holbrookia tex- ana. Herpetologica 4(6):223.
1949. Hibernation and the effect of a flood on
Holbrookia texana. Herpetologica 5(6): 125-126.
Richardson, C. H. 1915. Reptiles of northwestern Ne- vada and adjacent territory. Proc. U.S. Nat. Mus. 48(2078): 403-435.
Robison, W. G., and W. W. Tanner. 1962. A com- parative study of the species of the genus Crota- phytus Holbrook (Iguanidae), BYU Sci. Bull., Biol. Ser. 2(1): 1-31.
Romer, a. S. 1956. Osteology of the reptiles. Univ. Chicago Press, Illinois, xxi, 772 pp., 248 figs.
Savage, J. M. 1958. The iguanid lizard genera Uro- saurus and Uta with remarks on related genera. Zoologica 4,3(2):41-54.
Schmidt, K. P. 1921. New species of North American lizards of the genera Holbrookia and Uta. Am.
56
Great Basin Naturalist
Vol. 37, No.
Mus. Novit. (22): 1-6.
1922. A review of the North American genus
of Uzards Holbrookia. Bull. Am. Mus. Nat. Hist. I6(12):709-725.
1953. A checklist of North American amphibi- ans and reptiles. American Society of Icthyolog- ists and Herpetologists. University of Chicago Press, pp. 1-280.
Schmidt, K. P., and C. M. Bogert. 1947. A new fringe-footed sand lizard from Coahuila, Mexico. Amer. Mus. Nov. No. 1339:1-9.
ScHUCHERT, C. 1935. Historical geology of the Anti- Uean-Caribbean region, or lands bordering the Gulf of Mexico and the Caribbean Sea. John Wiley and Sons, New York, xxvi and 811 pp.
Simpson, G. G. 1945. The principles of classification and a classification of mammals. Bull. Amer. Mus. Nat. Hist., vol. 85, 811 pp.
Smith, H. M. 1935. Notes on some Mexican lizards of the genus Holbrookia with the description of a new species. Kansas Univ. Sci. Bull. 22(8): 185- 201.
1943. The White Sands earless lizard. Zool.
Ser. Field Mus. Nat. Hist. 24(30):339-.344.
1946. Handbook of lizards: Lizards of the
United States and Canada. Comstock Publ. Co., Ithaca, New York.
1960. Evolution of chordate structure. Holt,
Rinehart, and Winston, Inc., New York. XIV, 529 pp., .357 figs.
Smith, H. M., and D. M. Cochran. 1956. Callisatirus draconoides rhodostictiis Cope revived for the western fringe-footed lizards, Callisatirus draco- noides gabbii Cope. Herpetologica 12(2): 153- 154.
Smith, P. W., and W. L. Burger. 1950. Herpetologic- al results of the University of Illinois field expe- dition, spring 1949. Trans. Kans. Acad. Sci. S3(2): 165-175.
Stebbins, R. 1943. Adaptations in the nasal passages
for sand burrowing in the saurian genus Uma.
Am. Nat. 77:38-52. 1944. Some aspects of the ecology of the igua-
nid genus Uma. Ecol. Monogr. 14(3):31 1-332,
Figs. 1-22. 1948. Nasal structures in lizards with reference
to olfaction and conditioning of inspired air.
Amer. Jour. Anat. 8.3(2): 183-222. 1954. Amphibians and reptiles of western
North America. McGraw-Hill Book Co., New
York. 1966. A field guide to western reptiles and
amphibians. Peterson Field Guide Series.
Houghton Mifflin Company. Stejneger, L. H. 1890. Reptiles of San Francisco
mountain region. North American Fauna.
(3):109-111. Tanner, W. W., and J. Khogh. 1975. Ecology of the
zebra-tailed lizard Callisaurus draconoides at
the Nevada Test Site. Herpetologica 31(3):302-
316. Troschel, F. 1852. Cophosaurus texanus neve Eide-
chesengattung aus Texas. Arch. Naturgesch.
16(1)185, pp. 388-394. Wells, P. V., and C. D. Jorgensen. 1964. Pleistocene
wood rat middens and climatic changes in Mo-
jave Desert: A record of juniper woodlands. Sci- ence 143(3611): 1171-1 174. Williams, K. L., and H. M. Smith. 1958. Range and
status of a Mexican earless lizard. Herpetologica
13(4):265-267, pi. 1. Williams, K. L., P. S. Chrapliwy, and H. M. Smith.
1959. A new fringe-footed lizard, Uma from
Mexico. Trans. Kansas Acad. Sci. 62(2): 166-172,
pis. 1-2. WiLLisTON, S. W. 1925. Osteology of the reptiles. Har- vard Univ. Press, Cambridge, xiii, 30 pp., 191
figs.
HYDROLOGICAL FEATURES OF A CALIFORNIA COASTAL FEN
Don C. Erman', Kenneth B. Robv% and Michael Eames'
Abstract.— A unique fen peatland, on the CaUfomia coast, depends on six streams for its supply of flowing, mineral-rich water. About 25 percent of the water is supplied by surface streams that had average concentrations of Ca"'""'" from 0.15 meq/1 to 0.32 meq/1 and of Mg+ + from 0.25 meq/1 to 0.47 meq/1. During the dry sum- mer period oxygen concentration and pH were lower in the central area than in the ten margins. Water flow near the margins maintains high 0-2, pH, and cation content. The fen ranges from 4 m to over 11 m in depth, but the thickest peat layer is made up of very liquid, unconsolidated peat. In one year, an estimated 19.3 metric tons of suspended sediment entered tne 38.6 ha fen, which was equivalent to a 42 kg/ha loss from the total wa- tershed.
Inglenook Fen near the coast of Fort Bragg, California, may be the southernmost example of a fen-type peatland on the Pa- cific coast and is the only recognized ex- ample on the California coast (Baker 1972). A fen is a physical land type that has dis- tinctive flora and vegetation. It is often con- cave in cross section, has strong inflows of mineral-rich waters, has near-neutral pH, contains high amounts of Ca and Mg ions, and is productive (Heinselman 1970). To emphasize the source of minerals and water, this land type is referred to now as a miner- otrophic (Heinselman 1970) or rheotrophic (Moore and Bellamy 1973) peatland.
Fens evolve naturally over time to be- come bogs. Bogs (ombrotrophic peatlands) are characterized as being isolated from mineral-rich water, often convex in cross section, acid in pH, and unproductive (Heinselman 1970). Numerous examples of bogs exist along the Pacific coast of North America (Rigg and Richardson 1938). Sever- al are within a few km of Inglenook Fen in the "pygmy forest" on the upper coastal terraces (Rigg 1933, Jenny et al. 1969).
The peatland type and rate of change from fen to bog is strongly influenced by certain critical or controlling factors. Water sources are the key element in peatland ev-
olution (Heinselman 1970, Moore and Bel- lamy 1973). Little information is available on the important hydrologic features of In- glenook Fen. Our objectives were to deter- mine the sources and amounts of inflowing water, the nutrient status of the water, and the physical-chemical conditions in the fen.
Study Area
Inglenook Fen was described by Baker (1972), who included a floral list and some chemical conditions. The fen was formed by sand dunes blocking the stream outlet of a small valley on the second coastal terrace. A small 1.11 ha pond (Sandhill Lake) exists at the western edge, and radiating outward, especially to the east, are typical zones of marsh-fen vegetation— floating aquatics, emergents, sedge fen, and woody plants. The fen is 38.6 ha, of which about 4.6 ha are dominated by Cyperaceae fen and Ca- lamogrostis-CypersLceae-Menyanthes fen (cf. Baker 1972). The remainder is pre- dominantly woody vegetation (fen carr).
Six streams drain five watersheds (A-E) into the fen (Fig. 1). The watersheds range from forest land (67 percent) and pasture to .some residential development in Area C and part of Area D.
'Department of Forestry and Conservation, University of California at Berkeley, Berkeley, California 94720.
'U.S. Forest Service, San Francisco, California.
'Department of Fisheries. State of Wa.shington, Olympia, Washington 98501.
57
58
Great Basin Naturalist
Vol. 37, No. 1
M
ETHODS AND MATERIALS
M>
Approximately twice a month from No- vember 1972 to November 1973, we meas- ured discharge, water temperature, pH, dis- solved oxygen, calcium and magnesium ion content, total suspended sediments, ortho- phosphate, and nitrate in the fen streams. Silica was measured occasionally. Aerobic limit, peat depth, O2, pH, and temperature were measured irregularly on the fen.
Flow was measured in two ways. The first involved catching stream flow in a con- tainer of predetermined volume. The time required to fill a container was measured with a stopwatch. All streams except A^^^, Aj,^, and the outlet were measured this way as they emptied from culverts that passed beneath Highway 1 (Fig. 1). Flows of other streams that did not empty from culverts were determined by measuring a section for width, average depth, length, and average
velocity. Velocity was determined by timing a float over a measured distance.
Water temperature was determined with a thermistor or mercury thermometer, and oxygen concentration was determined with a YSI Model 51A Og-temperature meter or by the Winkler method.
Small samples were taken for water pH determinations. Shortly after collection of the sample, pH was measured with a La- Motte colorimetric comparator (wide range pH 3-10, narrow range pH 5-8).
Other water samples were taken for Ca- Mg ions and PO4-NO3 analysis. Samples for Ca-Mg analysis were preserved by addition of perchloric acid, filtered qualitatively be- fore analysis, and brought to a concentra- tion of 0.5 meq/1 of strontium chloride to reduce interference. Analysis was performed on a Perkin-Elmer Model 303 atomic ab- sorption flame spectrophotometer.
Separate glass bottles were used for PO4- NO3 samples; they were stored for one to three weeks in a refrigerator prior to analy- sis. Determinations were made on a Hach #640 Direct Reading Colorimeter. Ortho- phosphate was measured by the Hach- Stannaver Method and nitrate (after nitrite adjustment) by the cadmium reduction method.
An additional water sample was vacuum- filtered through a preweighed glass fiber fil- ter (approximately 0.3 jum pore size). The filter was then oven dried and reweighed to determine total suspended sediments.
Silica was measured with a LaMotte col- orimetric comparator. The depth to which oxygen was present in the fen (aerobic lim- it) was determined by measuring the stain that developed on implanted redwood stakes (Erman 1973). The distance from fen surface to where staining begins (absence of oxygen causes stain) is the aerobic limit. To determine peat depth, we made cores along two transects, A and B (Fig. 1), of the fen with a Hiller-type peat borer.
Fig. 1. Map of Inglenook Fen showing inlet streams (dashed hnes) and their watersheds (A to E). Lines a and b are transects for peat depth cores. The dotted area in the central fen region is predominantly sedge vegetation.
Results and Discussion
Stream Flows
Mean surface flows are given in Table 1. Large seasonal fluctuations occurred in dis-
March 1977
Erman, et al.: California Coastal Fen
59
Table L Yearly average (and standard deviations) of environmental conditions of outlet and inlets to Inglenook Fen, Mendocino Co., California.
|
Environmental Factor |
Aso |
Ano |
Stream B |
C |
D |
E |
Outlet |
|
Discharge (1/sec) (S.D.) |
4.93 (3.97) |
3.26 (4.93) |
14.6 (14.1) |
9.17 (17.7) |
9.03 (14.8) |
4.25 (11.5) |
179.8 (278) |
|
0-2 (mg/1) (S.D.) |
10.1 (1.24) |
10.0 (2.13) |
10.5 (0.94) |
10.1 (1.31) |
11.3 (1.66) |
10.3 (1.01) |
7.43 (1.92) |
|
O2 saturation (%) (S.D.) |
94.4 (9.47) |
89.3 (16.8) |
101.2 (5.56) |
95.1 (6.3) |
101.5 (10.7) |
95.8 (7.64) |
70.5 (24.4) |
|
Temperature (°C) (S.D.) |
12.7 (3.3) |
12.0 (3.4) |
12.8 (3.7) |
12.7 (4.4) |
113 (4.0) |
12.9 (4.9) |
13.4 (5.3) |
|
Median pH |
6.5 |
5.6 |
6.7 |
6.8 |
6.8 |
6.3 |
6.6 |
|
Ca+2 (me/1) (S.D.) |
.16 (.09) |
.14 (.08) |
.18 (.08) |
.28 (.14) |
.32 (.15) |
.15 (.07) |
.62 (.25) |
|
Mg+2 (me/1) (S.D.) |
.27 (.06) |
.26 (.06) |
.33 (.09) |
.47 (.14) |
.47 (.14) |
.25 (.06) |
.47 (.10) |
|
Sediments (mg/1) (S.D.) |
19.9 (22.9) |
11.4 (12.6) |
16.0 (20.8) |
12.5 (11.6) |
16.9 (18.6) |
4.24 (3.50) |
4.41 (5.59) |
|
PO, -P (mg/1) (S.D.) |
.09 (.08) |
.09 (.07) |
.08 (.05) |
.09 (.08) |
.09 (.08) |
.07 (.07) |
.13 (.08) |
|
NO3 - N (mg/1) (S.D.) |
.60 (.29) |
.23 (.26) |
.28 (.07) |
.13 (.07) |
.28 (.16) |
.06 (.03) |
.12 (.10) |
|
Conductivity'' (uMhos @ 25°C) |
109 |
216 |
154 |
222 |
203 |
92 |
.366 |
^Data from November only.
charge with minimum flows from May to October. Stream B consistently had the highest flow (avg. 14.6 1/sec). The total out- flow from May to October (the dry season) averaged 47.6 1/sec (28-68 1/sec) and was more than twice the combined inflow (15.9 1/sec). Mean monthly discharge of the com- bined inlets and of the outlet are shown in Fig. 2.
One of the most interesting results of this work is the relationship between inlet and outlet flows. R. Jackson (pers. comm.), who surveyed the fen for the Nature Con- servancy, surmised that the predominantly sandy soils of the watershed would result in much subsurface flow. For the entire sam- pling period the measurable surface inflow
averaged only 25.2 percent (13.8-66.3 per- cent) of the measurable surface outflow. With one exception, during a rainstorm on 21 September 1973, measurable inflow nev- er exceeded 50 percent of outflow. Thus subsurface flow appears more important than surface flow in this area. Water, of course, also enters the fen from surrounding sand dunes and direct precipitation on the fen, and it leaves via evaporation and evap- otranspiration. These sources were not measured.
By expanding the mean discharge value we estimated the total annual volume of water carried by each stream (Table 2). The estimated yearly inflow of all surface water was 1.43 million m^, compared to about
60
Great Basin Naturalist
Vol. 37, No. 1
|
1549 |
K4a |
||||
|
/ |
\ FEN OUTLET |
||||
|
500 |
• |
\ |
|||
|
ro" O |
. |
\ |
|||
|
'^5 400 |
■ |
"^SURFACE V |
|||
|
Hi |
\ STREAMS \ |
||||
|
DISCHAR( |
:\^ |
\ \ |
; |
||
|
200 |
- |
\ \ |
, / |
||
|
Z o 2 100 |
: |
/ |
\ ^ |
__^ |
■^ |
|
. |
^^"--^j_^_ |
^______. |
^^^' |
||
|
, |
1 1 1 1 1 |
• |
NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT 1972 1973
DATE
Fig. 2. Estimated total discharge by month of all inlet streams combined and of the outlet.
5.67 million m^ of outflow. In Table 2 the discharge of each inflowing stream is shown as a percentage of the total inflow, and this value is compared with the percentage of the total watershed area for each of the streams. These results indicate differences between size of watershed and amount of surface flow. For example, based on surface flow. Watershed A accounted for 18.1 per- cent of the water volume while it included 42.7 percent of the total watershed. Water- shed B contained 12.4 percent of the area, but accounted for .32.3 percent of the total
flow. Differences shown in these figures may be due to differences in vegetative cover, soil type, and slope gradient in the five wa- tersheds. Preliminary soil maps of the fen watershed show streams A to C drain Em- pire Sandy Loam, while streams D to E drain Hugo Sandy Loam. We have no data on subsurface flow, but it may be that when subsurface flow is added to surface flow, the total contribution of each stream is closer to its percentage of the total watershed.
Mean monthly discharge of the combined inlets and of the outlet are shown in Fig. 2.
Table 2. Comparison of surface stream inflow and area of watershed of Inglenook Fen.
|
Total |
Area of |
|||
|
Annual Flow |
% of Total |
Watershed |
% of Total |
|
|
Stream |
(M3 X 103) |
Flow |
(Hectares) |
Watershed Area |
|
A (No -(- So) |
258 |
18.1 |
196.3 |
42.7 |
|
B |
460 |
32.3 |
86.7 |
12.4 |
|
c |
289 |
20.2 |
33.8 |
18.3 |
|
D |
285 |
20.0 |
77.7 |
16.9 |
|
E |
134 |
9.4 |
44.9 |
1.7 |
Total
1430
459.4
March 1977
Erman, et al.: California Coastal Fen
61
Jackson (pers. comm.) has suggested that only Stream E is seasonal. Stream A^^, was the only intermittent stream during the pre- sent study, although Stream E carried less than 0.3 1/sec from June to September.
Temperature and Oxygen
It was impossible to sample the streams at the same time of day on the various dates. Daily fluctuations in temperature are unknown for these streams. As expected, the winter water temperature (5-12 C) was lower than the summer water temperature (15-20 C), although temperatures below 10 C are unusual. High daytime air temper- atures are uncommon in summer because of frequent fog along the coast, and thus even small streams are not particularly warm.
In general, oxygen concentration in inlet streams was near air saturation (average 95- 101 percent except for A^^^, which had low O2 (5.7-7.3 mg/1) at low flows (Table 1). Warmer water temperatures during low flows resulted in lower oxygen concentra- tions, as shown for Stream B in Fig. 3, than at other times of the year, although oxygen
was still near air saturation. For most of the year, the outlet had significantly lower Og content than the inlets. The outlet is com- pared to Stream B in Fig. 3. This result is expected because the fen has a tremendous amount of decaying material that would consume oxygen as water flowed through its length. Much of the inflow water also even- tually passes through the small lake near the outlet. Lake O2 levels were consistently below saturation, especially at lower depths. The outlet stream generally was below 60 percent saturation (less than 7 mg/1) during the wet months; but from June to Septem- ber, when outflows were less than 50 1/sec, the oxygen levels were at their highest (89- 111 percent saturation, 8.4-11.2 mg/1). Higher oxygen content in the outlet during low water periods probably reflects less contact of inflowing water with fen peats. During the winter months water is ob- viously moving across much of the fen sur- face. But when inflows diminish to some minimum level (probably at least 50 1/sec combined surface and subsurface), the water tends to move only through the marginal channels ("moats," e.g. Rigg 1940) of the
10-
6
2 UJ
X ^ 0
— INLET STREAM B —OUTLET STREAM
pH
NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV
1972
1973
□ATE
Fig. 3. Seasonal changes in oxygen concentration and pH in the outlet compared to the major inlet stream (B).
62
Great Basin Naturalist
Vol. 37, No. 1
fen. We observed that the central sedge area was noticeably drier during the period from June to September, and this observa- tion supports the idea of the sedge area's reduced contact with flowing water and of lower water levels. As long as this oxygen- rich (and relatively mineral-rich) water flows through the central fen, the rate of peat accumulation will be slower than if water flow is confined to the margins. As peat and living plants in the center of a fen become isolated from flowing water, charac- teristic changes occur (Gorham 1957, Deevey 1958, Heinselman 1970, Moore and Bellamy 1973): the rate of peat accumula- tion accelerates with a consequent rising of the surface; and the peaty soils become pro- gressively less rich in nutrients, more acid in pH and more bog than fen. Recent work in Manitoba, Canada, showed that about 74 percent of the net primary production of a bog remained after one year, whereas 52 percent remained in a marginal fen (Reader and Stewart 1972). Changes in the biota ac- company such changes in physical-chemical conditions. There is, then, a very important influence of the quantity of water and its oxygen and mineral conditions on the rate of succession of Inglenook Fen.
Acidity (pH)
Median pH of the inlet and the outlet streams is given in Table 1. There was a general trend of streams with lower dis- charge to have lower pH. From May through August, when inflows were derived almost entirely from aquafer rather than run-off, pH was generally above neutral (7.0-7.7) or, in the case of Aj^^, and E, was higher than at other periods, although it was less than 7.0. In contrast, streams had lower pH in months when flows were high.
Tests of oxygen, temperature, and pH in the fen were made generally along Transect A. Average values from these collections are given in Table 3. Samples were taken near the fen edge, where moving water occurred all year, as well as toward the center, where water was low in summer and had little flow. These average values showed seasonal changes in oxygen, temperature.
Table 3. Average selected chemical conditions Inglenook Fen.
|
Date |
Number of |
Oxygen |
Temp. |
pH |
|
Samples |
(mg/1) |
(°C) |
||
|
Dec. 2, 1972 |
7 |
4.9 |
9.2 |
6.0 |
|
Jan. 1, 1973 |
4 |
5.8 |
7.1 |
— |
|
Feb. 10 |
4 |
5.8 |
11.2 |
6.2 |
|
Feb. 24 |
4 |
8.7 |
10.0 |
6.2 |
|
Mar. 24 |
6 |
6.2 |
13.9 |
6.0 |
|
Apr. 7 |
6 |
3.5 |
16.8 |
6.0 |
|
Apr. 21 |
5 |
3.5 |
16.8 |
5.9 |
|
May 10 |
3 |
2.6 |
13.3 |
6.0 |
|
May 19 |
3 |
3.0 |
17.2 |
5.8 |
|
June 19 |
8 |
2.5 |
14.6 |
5.8 |
|
July 2 |
3 |
3.3 |
16.0 |
5.7 |
|
July 14 |
6 |
0.7 |
18.3 |
5.8 |
|
Aug. 1 |
6 |
.3.0 |
18.0 |
5.9 |
|
Aug. 13 |
2 |
0.6 |
18.7 |
5.6 |
|
Sept. 4 |
5 |
3.0 |
18.8 |
5.6 |
|
Sept. 21 |
3 |
3.4 |
20.0 |
6.6 |
|
Nov. 3 |
5 |
5.0 |
9.2 |
5.8 |
and pH that were the opposite of the in- coming streams [see Stream B in Fig. 3). In winter months high water flushes the entire fen and elevates oxygen and pH (especially in the central region). During drier periods, oxygen and pH decline in the fen except near the marginal channels. Oxygen, pH, and aerobic limit (to be discussed later) are increased relatively close to a region of moving water (Lahde 1969, Heinselman 1970, Erman 1973). Recent studies on a raised bog-system in Germany (Lotschert and Gies 1973, Gies and Lotschert 1973) showed that the marginal fen areas were consistently higher in pH and cation con- tent than the raised, central portion of the peatland.
The pH of the outflow stream also in- dicates isolation of water from the central fen region during dry periods. Instead of pH decreasing— a condition that would oc- cur if the water were in contact with the more acid central area— the pH of the out- flow increases (Fig. 3). The studies by Lot- schert and Gies (1973) also showed a sum- mer increase in pH and cations in the marginal fen of Schwarze Moor. This change in pH in Inglenook Fen could result from additions of water that drain the alka- line dunes and that are less diluted by stream inflows at this time.
March 1977
Erman, et al.: California Coastal Fen
63
Gorham (1975) suggests a pH of 4.2 as a lower limit of fen conditions, and Lotschert and Gies (1973) give ranges of 4.9 to 6.3 for typical fen conditions. Previous work (Baker 1972) and our present samples from the sedge-fen region indicate pH levels of 5.4 to 6.6, depending on season (more acidic dur- ing dry periods).
Calcium-Magnesium ions
Calcium ions averaged 0.15 meq/liter (3 mg/1) at Stream E to 0.32 meq/1 (6.4 mg/1) at Stream D (Table 1). These values are lower than observed on one occasion by Baker (1972), who found 0.40 meq/1 in Stream E. The outlet was consistently high- er in Ca ions (average 0.62 meq/1) than the inlets, and probably reflects the influence of dunes that were found to have higher Ca ion concentration than inflow water (Baker 1972). As a further check on the possibility of the dunes influencing the outlet water, a test for silica was made 7 October 1973. If dune water is important, then silica should be much higher in the outlet than in the in- let streams. The average for the six inlets was 3.5 mg/1, compared to 8.5 mg/1 for the outlet. These results further indicate the rel- ative importance of direct drainage from the dunes into the fen margins.
Magnesium ions varied similarly to cal- cium in all inlet streams (Table 1), and ranged from 0.25 meq/1 (E) to 0.47 meq/1 (C and D) or 3.5 to 5.7 mg/1. The outlet, however, contained about the same as the inlets (0.47 meq/1), which is in contrast to the situation for calcium as noted previously by Baker (1972).
Sea water spraying directly onto the fen may also contribute to the concentration of various ions in the