North American Fauna, No. 14. 3 9 d .- o •£ * £ HI IE .5 If 2 I I W I 8 « e ui C o "^ Q. g U. S. DEPARTMENT OF AGRICULTURE DIVISION OF BIOLOGICAL SURVEY NORTH AMERICAN FAUNA USTo. 14 [Actual date of publication, April 29, 1899] NATURAL HISTORY OF THE TRES MARIAS ISLANDS, MEXICO General Account of the Islands, with Reports on Mammals and Birds. By E. W. NELSON Reptiles of the Tres Marias. By LEONHARD STEJNEGER Notes on Crustacea of the Tres Marias. By MARY J. RATHBUN Plants of the Tres Marias. By J. N. ROSE Bibliography of the Tres Marias. By E. W. NELSON Prepared under tlio direction of Dr. C. HART MERRIAM CHIEF OK DIVISION OF MOLOGICAL SURVEY WASHINGTON GOVERNMENT PRINTING OFFICE 1899 LETTER OF TRANSMITTAL. DIVISION OF BIOLOGICAL SURVEY, Washington, D. C., January 25, 1899. SIR: I have the honor to transmit herewith for publication as North American Fauna No. 14 a report by E. W. Nelson on the natural his- tory of the Tres Marias Islands, Mexico. These islands are the largest off the west coast between Cape St. Lucas and the Isthmus of Panama, but have seldom been visited, and very little is known of their fauna or flora. For several years Mr. Nelson has had charge of the field work of the Biological Survey in Mexico, and in May, 1897, visited the Tres Marias. During the course of this visit he made a thorough collection of birds and mammals and also secured specimens of reptiles, fishes, mollusks, crustaceans, and plants, so that his report contains a fairly complete account of the natural history of the islands. In work- ing up the material collected, Mr. Nelson has had the assistance of several well known naturalists in the United States National Museum and United States Fish Commission, who have prepared reports on special groups, as credited in detail on page 13. Mention should be made also of the unfailing courtesy and interest of the Mexican Government in the investigations conducted by the Biological Survey in Mexico. Letters have been furnished by officials in the City of Mexico, and by the late Mexican minister in Washing- ton, Sefior Don Matias Romero, which greatly facilitated the work in various ways, and on the occasion of the visit to the Tres Marias enabled Mr. Nelson to borrow a large boat at San Bias and secure comfortable quarters on the islands. Several attempts at agriculture have been made on the Tres Marias Islands, but the results have thus far been unsuccessful, owing to the dry climate and the scarcity of permanent water. Corn and beans have been grown on a small scale, but the crops suffer from the severe storms which occur at certain seasons. Experiments have been made with a view to utilizing the native species of agave for fiber and mes- cal, and the cultivation of cotton has also been tried without success. Recently it has been proposed to establish an American colony on one of the islands for the purpose of growing coffee, bananas, Australian 4 LETTER OF TRANSMITTAL. chestnuts, and date palms, and to engage in the manufacture of banana and chestnut flour. Such a scheme, Mr. Nelson tells me, could only result in failure, as the islands are entirely unsuited to growing these products. It therefore seerns desirable to publish at once all the information in the possession of the Department, for the purpose of making it available to those who may be interested in the islands or their products. Eespectfully, C. HART MERRIAM, Chief, Biological Survey. Hon. JAMES WILSON, Secretary of Agriculture. CONTENTS. Page. General description of the Tres Marias Islands, Mexico. By E. W. Nelson 7 Mammals of the Tres Marias Islands. By E. W. Nelson 15 Birds of the Tres Marias Islands. By E. W. Nelsou 21 Reptiles of the Tres Marias Islands. By Leonhard Stejneger 63 Notes 011 Crustacea of the Tres Marias Islands. By Mary J. Rathbun 73 Plants of the Tres Marias Islands. By J. N. Rose 77 Partial Bibliography of the Tres Marias Islands. By E. W. Nelson 93 ILLUSTRATIONS. PLATE. Map of the Tres Marias Islands Frontispiece. FIGURES. Page. 1. Erythrina lanata Rose 81 2. Euphorbia nelsoni Millspaugh 89 5 No, 14, NORTH AMERICAN FAUNA, April 29, 1899, GENERAL DESCRIPTION OF THE TRES MARIAS ISLANDS, MEXICO. By E. W. NELSON. INTRODUCTION. The Tres Marias islands are situated off the west coast of Mexico, about C5 miles west from the port of San Bias. These islands have been known since early in the history of the ^New World, and in 1532 were named Las Islas de la Magdalena by Diego de Mendo/a. Many of the early explorers sailed about them, and Dampier states that they were familiar to the buccaneers who visited these shores. They are mentioned by several of the later voyagers, especially the English exploring expeditious which visited the west coast of Mexico in 'the first half of the present century. During all this time, however, they remained uninhabited and nothing definite was known or published concerning their character or products. It is said at San Bias that the first men who lived upon the islands were bandits, who took refuge there, and had a secure retreat from which they harried the mainland settlements for several years. Finally, the abundance of Spanish cedar became known, a settlement of woodcutters was established on Marie Madre, and this island has since been continuously inhabited. Col. A. J. Grayson, a naturalist who lived for many years on the west coast of Mexico, was the first to publish any detailed information about the islands.' Most of this information is contained in the various papers embodying the results of his three trips to the Tres Marias in 18G5, 'G6, and '67, published by himself, George N. Lawrence, and W. E. Bryant. In 1881 Alphouse Forrer, a natural history collector, spent some time on Maria Madre collecting specimens for the British Museum, but no de- tailed account of his work has been published. No other naturalist is known to have visited the islands until the spring of 1897. In April of that year Mr. E. A. Goldman and I visited the port of San Bias for the 1 Mr. John Xantus, who spent several years subsequent to 1859 on the west coast of Mexico, was supposed to have visited the islands, on account of several specimens of birds which ho sent to the Smithsonian Institution, labeled " Tros Marias Islands, 1861." But as no one else has collected any of these species, and as Xantus sent in no birds which have been taken by others on the islands, it is safe to conclude that he did not visit the Tres Marias. 7 8 NORTH AMERICAN FAUNA. purpose of outfitting an expedition to the Tres Marias. A letter to the collector of customs at San Bias, kindly furnished me by the Mexican Minister in Washington, the late Don Matias Romero, proved of the greatest service. The collector of customs rendered every assistance in his power, including the loan of a large open boat 25 feet long, and a letter to his deputy which secured us very pleasant quarters in the custom house on Maria Madre. While preparations for the trip were in progress a party from Socorro, ST. Mex., consisting of Prof. C. L. Herrick, his son Harry, and Dr. T. S. Maltby arrived at San Bias, also bound for the Tres Marias, and we made the trip together. On the evening of April 28 the boat crept out of the lagoon, and by the aid of a faint land breeze edged slowly off shore. The islands came in sight the next morning, but it was impossible to reach them for several days, owing to calms, head winds, and the lack of a keel to the boat. The stock of water was on the point of exhaustion when Maria Madre was finally reached, three days later, on the afternoon of May 2. The landing was made at the settlement at the head of a shallow bay on the east side of the island. Our letters secured a cordial welcome from the customs inspector and the agent of the owner of the islands. In a couple of hours the outfit was snugly installed on the broad upper verandas of the custom house, where our headquarters were located. Collections were made near this place, the island trav- ersed both on foot and horseback, and on May 20 a boat trip was made to the north end of the island and across to San Jnanito. On May 23 the party returned to the settlement, and two days later proceeded to Maria Magdalena, where camp was made near the beach for four days. On May 29 we crossed to Maria Cleofa, where we remained two days, and then started, May 31, on the return to the mainland. The wind was fair, and a quick trip was made, San Bias being reached on the evening of June 1. When Colonel Grayson visited the islands, in 1865, he found a settle- ment on Maria Madre, but the other islands uninhabited. In the spring of 1897 there was a branch custom-house, with three inspectors, at the main settlement on Maria Madre, which had supervision of the shipment of salt and Spanish cedar. The settlement contained about twenty-five families, all of whom, except the customs inspectors, were in the service of the owner of the islands, Sefiora Gil de Azcona, who lived in the city of Tepic, on the mainland. In May and June the workmen are employed in salt-making at a lagoon near the south point of the island, where there is a small group of houses. The rest of the year they are occupied in cutting cedar and hauling it to the beach for shipment. The available supply of this valuable timber is now approaching exhaustion. Subsequent to Grayson's visit a settle- ment of woodcutters was made on the northeast side of Maria Magda- lena, and a number of houses were built and a field cleared. We found the place deserted, the houses in ruins, and the field overgrown with thorny bushes. GENERAL DESCRIPTION OF THE TRES MARIAS ISLANDS. 9 The amount of land suitable for agriculture upon the islands is very limited and forms but a small percentage of the total area. A few cattle are raised on Maria Madre, but the scanty herbage and great scarcity of water during the long dry season limit this industry to the most insignificant proportions. There is a small field near the settle- ment, where coarse grass is grown for stock. Attempts have been made to grow corn and beans to supply the residents, but the fierce summer storms of wind and rain, called 'chubascos,' which beat the crops to the ground, have rendered these efforts futile. At present all food supplies are brought from the mainland. A number of years ago a house was built and a field cleared and fenced near the north end of the island for the purpose of growing cotton. A warehouse was also built at the main settlement, but after a trial the owner was forced to abandon the industry, the field and house were deserted, and the place is now overgrown with bushes. Subsequently it was proposed to utilize the agaves, which grow abundantly near the north end of the island, for fiber and for distilling from their fleshy bases the alcoholic product known as 'mescal.' Machinery was obtained, but the owner died before the industry was exploited. In winter the weather is dry and pleasant, and small coasting steam- ers stop every now and then to take on wood for fuel, and sailing vessels call for Spanish cedar or, in spring, for salt. In May the inhabitants are obliged to lay in a stock of provisions sufficient for several months, as they are practically cut off from communication with the mainland during summer, when the islands are avoided on account of the storms that sweep over them. Many objects drift out from the Gulf of California after storms and are cast up on the shores. In September, 1896, a great tornado of wind and rain swept over northern Sinaloa and the Gulf of California; the coast lowlands were devastated by the flooded rivers, and crops and forests were alike overwhelmed and swept to sea. In May, 1897j the shore of Maria Madre was still strewn with cornstalks, driftwood, and other wreckage that had been stranded after this storm. Our obligations to the collector of customs at San Bias and his agent on Maria Madre have already been mentioned, and acknowledgments are due also to the owner of the islands, Sefiora Gil de Azcona, whose letter procured us the use of horses and other courtesies. PHYSIOGRAPHY. The Tres Marias are situated between latitude 21° and 22° and longitude 106° and 107° (see frontispiece). Between the islands and the mainland, 20 miles offshore, lies Isabel Island, only about a mile long and 150 feet high. The soundings in the channel between the mainland and the islands gradually deepen to less than 300 fathoms, but just west of the group the sea bottom drops rapidly to more than 1,500 fathoms. The absence of a deep channel shows that they are continental islands, as distinguished from thg£tf0&aHic lievillagigedo group, farther west. - 10 NORTH AMERICAN FAUNA. The Tres Marias group comprises four islands, San Juanito, Maria Madre, Maria Magdalena, aud Maria Cleofa, arranged in a northwest and southeast direction. Maria Madre, the largest, measures about 8 by 15 miles, aud rises over 2,000 feet above the sea. North of this, and separated from it by a channel 4 miles wide and 5 or 6 fathoms deep, is San Juanito, an islet 3 or 4 miles in diameter and about 100 feet high. Next southeast of Maria Madre is Maria Magdalena, roughly triangular in outline and 7 or 8 miles across, with its central summit rising to an altitude of about 1,500 feet. A shallow channel 8 miles wide separates it from Maria Madre. Southeast of Maria Mag- dalena lies Maria Cleofa, the last of the group. It is irregularly rounded in outline, about 3 miles across, and its altitude is apparently much less than 1,320 feet, as given on the charts. The channel between the two last-named islands is about 12 miles wide and much deeper than the others. With the exception of San Juanito, which is nearly flat with a narrow border of low bluffs along the north shore, the islands are mountainous and rise in successive slopes from the shore to the culminating point near the center. The interior of Maria Madre is occupied by a moun- tainous ridge extending almost the entire length of the island, but descending to a gently sloping area near each end. The eastern side of the island has the longer slope, while the westward or seaward face is much more abrupt, thus corresponding with the formation of the mountains parallel to the coast on the adjacent mainland. Both slopes of the island are scored at intervals with canyons which usually descend in a nearly direct line to the sea. Maria Magdalena and Maria Cleofa are occupied by a central mountainous elevation, from which canyons descend in all directions to the sea. The northeastern points of both these islands are low, flat, sandy areas of limited extent, and the west- ern faces are rocky and precipitous. Permanent fresh water is very scarce on all the islands. There are three little streams on Maria Madre, which sink several miles from the sea during the dry season, and one each on Maria Magdalena and Maria Cleofa. The relative situation of the islands, with the narrow, shallow chan- nels between them, shows conclusively that at one time they formed a single island at least 45 or 50 miles long, and at a still earlier stage they must have been connected with the mainland. One of the strong- est proofs of this former connection is shown by the correspondence between the fauna and flora. The breaking down of the original island into several smaller ones and the evident continuous encroachment of the sea appear to indicate that the subsidence is still in progress. The country back of the coast on the mainland was, within a comparatively recent period, the scene of great volcanic activity, and the Tres Marias bear evidence of having undergone various oscillations in level. On Maria Madre there are great beds of marine deposits, hundreds of feet above sea level, containing quantities of shells and corals of species now living along the shore. Isabel Island, near the mainland, is of GENERAL DESCRIPTION OF THE TRES MARIAS ISLANDS. 11 volcanic origin and exhibits similar evidence of having once been a much larger island which is now sinking. Apparently it consists mainly of the remains of an old volcano, and a small crater still occupies the center of the island. Although no craters were seen on the Tres Marias, yet there are lavas and other volcanic rocks on all the islands, but a large part of the formation is made up of other rocks elevated by the volcanic uplift. FAUNA. The Tres Marias, like the adjacent coast, lie within the Arid Tropical life zone. The evidence furnished by the fauna of the former connec- tion of the Tres Marias with the mainland is as follows : Six species of land shells were obtained, which, according to Dr. William H. Ball, are widely distributed on the mainland. These species are Polygyra ventrosula Pfr., Orthalicus undatus Brug., Orthalicus undatm melano- cheilus Val., Lamellaxis ?, Opeas subula Pfr., and Glandina turris, Pfr. A fresh- water fish taken on Maria Magdalena and Maria Cleofa has been identified by Prof. B. W. Evermann as Agonostomus nasutus Giinther, a common species on the mainland. In fresh-water pools on Maria Magdalena two or three individuals of another small fish were seen, which were very similiar to common mainland species of Aicaous, and undoubtedly belong to this or a closely allied genus. Six of the seven species of lizards inhabit the mainland, and only one is peculiar to the islands; the mud turtle and crocodile are also found on the mainland, as are the eight species of snakes. Concerning the reptilian fauna Dr. Stejneger remarks: "Thus most of the species are common on the opposite mainland and generally distributed over tropical Mexico and Central America. Then again it seems as if the species are prac- tically identical on all the islands of the group. This would indicate a comparatively recent severance of these islands from each other, as well as from the opposite mainland of Mexico." The birds and mammals seem to have been more susceptible to modi- fying influences than other forms of life. Thirty-six species of resident land birds were found on the group, of which twelve are identical with those on the mainland, and twenty-four can be distinguished specifically or subspecifically. We found ten species of indigenous mammals, seven of which, acccording to Dr. Merriam, are peculiar to the islands, but closely related to species living on the mainland. ANIMALS PECULIAR TO THE TRES MARIAS. So far as known, the following species and subspecies (with the excep- tion of Compsothlypis insularis) are peculiar to the islands: Marmosa insularis Merriam. Procyon lotor insularis Merriam. Oryzomys nelsoni Merriain. Bhoyeessa parvula H. Allen. Peromyscus madrensis Morriam. Glossophaga mutica Merriam. Lepus grayeoni Allen. 12 NORTH AMERICAN FAUNA. Columba flavirostris madrensis Nelson. Leptotila capitalis Nelson. Buteo borealis fumosus Nelson. Polyborus cheriway pallidits Nelson. Psiltacula insularis Ridgway. Trogon ambiguus gold-man i Nelson. Dryobates scalaris graysoni Baird. Nyctidromus albicollis insularis Nelson. Amazilia yraysoni Lawrence. lache lawrencei Ridgway. Platypsaris aglaice insularis (Ridgway). Myiopagis placens minimus Nelson. Icterus graysoni Cassin. Cardinalis cardinalis marice Nelson. Piranga bidentata flammea (Ridgway). Vlreo flavoriridis forreri (Von Madarasz). Vireo hypochryseus aordidus Nelson. Compsotlilypis insularis (Lawrence). Oc- curs also on the mainland near San Bias. GranateUus francescce, Baird. Thryothorus lawrendi (Ridgway). Thryothorus lawrendi magdalence Nelson. Melanotis ccerulescens longirostris Nelson. Myadestes obscurus insularis Stejneger. Merula graysoni Ridgway. Cncmidophorus mariarum Giinther. FLORA. The islands were visited near the end of the long dry season, when most of the herbaceous plants were withered and lifeless, but repre- sentatives of 136 species, largely shrubs and trees, were secured. The general appearance of the vegetation was the same as that in similar situations on the mainland. Among the most notable plants were the Spanish cedar (Cedrela), three species of wild fig (Ficus), two of Pithecolobium, five of Solanum, two of Ipomcea, a Passiflora, cassias, euphorbias, a large agave, a large cereus, and two opuntias. On San Juanito the vegetation is largely made up of bushes and scrubby trees 8 to 15 feet high, with many agaves on the sandy southern end. Agaves are very numerous also on the northern end of Maria Madre. On the latter island the forest is rather low and scrubby near the shore, but increases in luxuriance farther up the slopes, especially along the bottoms and sides of the canyons, where Spanish cedars, wild figs, and several other trees attain a large size. In its primeval con- dition, before the advent of woodcutters, it must have presented a fine example of tropical forest growth. Now, only a few specimens remain to show what the original condition must have been. Along the sum- mit of the island the dense forest is made up of slender-trunked trees, called 'palo prieto' by the natives, which I was unable to identify. On Maria Magdalena the conditions were similar to those on Maria Madre, but a larger percentage of the original forest still remains intact, although the Spanish cedars are mainly gone. Maria Cleofa is more rocky and sterile, and the trees are stunted and brushy. Several spe- cies found on the other islands appeared to be wanting here. The report on the plants shows that the flora of the islands is very similar to that of the mainland, and the fact that several new species were found may be due to our imperfect knowledge of the mainland flora. GENERAL DESCRIPTION OF THE TRES MARIAS ISLANDS. 13 PLANTS DESCRIBED FROM THE TRES MARIAS. JEgiphila pacifica Greeiiman. Seloperone nelsoni Greenmaii. Buxus pubescens Greenman. Cordia insularis Greenman. Gilibertia insularis Rose sp. nov. Pilocarpita insularis Rose sp. nov. Ternostroemia maltbya Rose sp. nov. (also on mainland). Erythrina lanata Rose sp. nov. (also on Zanthoxylum insularis Rose sp. nov. mainland). Euphorbia nelsoni Millspaugh. Euphorbia subccerulea tresmariw Millsp. var. nov. Zanthoxylum nelsoni Rose sp. nov. SUMMARY. The following statement shows the number of species of animals and plants now known from the Tres Marias : Land mammals 11 Birds 83 Reptiles 18 Fresh- water fish... 2 Fresh- water shrimp 1 Land mollusks 6 Plants ..136 ACKNOWLEDGMENTS. Much of the value of this report is due to the cordial cooperation of several eminent specialists. Through the courtesy of Mr. F. V. Coville, curator of the National Herbarium, Dr. J. N. Bose, assistant curator, was enabled to prepare the report on the plants. Dr. Leonhard Stej- neger, curator of the division of reptiles of the National Museum, Dr. William H. Dall, honorary curator of the division of conchology, and Miss Mary J. Eathbun, assistant in the division of invertebrates, reported on the Tres Marias material; and Prof. B. W. Evermanu, ichthyologist of the United States Fish Commission, kindly identified the collection of fishes from the islands and the adjacent mainland. Finally, I wish to express my great indebtedness to Mr. Robert Eidg- way, curator, and Dr. Charles W. Eichmond, assistant curator, of the division of birds in the National Museum, for having so freely placed at my disposal, not only the material in their charge but also their knowledge of tropical American birds. MAMMALS OF THE TRES MARIAS ISLANDS. By E. W. NELSON. Mammals are not numerous either in species or individuals upon the Tres Marias. So far as known, they number but eleven species, of which seven are peculiar to the islands; one is introduced, and the other three are widely ranging bats. A sea lion and two species of porpoise were found near the shores, and whales were reported to occur during certain seasons. As with the birds, one of the most unaccountable features of the mammal fauna is the absence of a num- ber of species that are common on the adjacent mainland. Considering the primitive condition of the islands, it is difficult to explain the presence of field mice, the pigmy opossum, rabbit, and raccoon, while the large gray opossum, nasua, skunk, fox, coyote, deer, peccary, squirrel, and various small rodents of the adjacent mainland remain unrepresented. The Tres Marias mouse was rather common above 200 feet on all of the larger islands; the rabbit was very numerous near the north end of Maria Madre, on San Juanito, and in some places on Maria Magdalena, and two species of bats were abundant in caves on Maria Madre. Aside from these species, mammals were uncommon and difficult to find. One cause of their general scarcity may be the very limited supply, of permanent fresh water, and the absence of small species from a broad belt near the shore was easily accounted for by the abundance of carnivorous crabs. The mammals obtained by our party have been identified by Dr. C. Hart Merriam, who has described the new forms and given critical notes on other species.1 Of the land mammals taken, five were new and two, Lepus graysoni and Rhogeessa parvula, had been previously described. We failed to secure two species of bats (Myotis nigricans and Lasiurus borealis mexicanus) which were taken by Mr. Forrer. Not- withstanding the fact that collections were made in several branches of natural history, I feel confident that representatives of all the resi- dent land mammals were secured, but it is quite possible that future work may add other bats to the present list. ANNOTATED LIST OF SPECIES. Marmosa insularis Merriam. Tres Marias Pigmy Opossum. Marmosa insularis Merriam. Proc. Biol. Soc., Washington, XII, pp. 14-15, Jan. 27, 1898. Type from Maria Madre Island. These pretty little opossums were not found except in the high inte- rior of Maria Madre, between 1,200 and 1,800 feet above sea level, where ' Proc. Biol. Soc. Washington, XII, pp. 13-19, 1898. 15 16 NORTH AMERICAN FAUNA. they were apparently rather common about the wild fig trees in the forest and were feeding upon the figs. They may occur also on the other islands, especially upon Maria Magdalena. Two men living on the island described the nests of these animals as globular masses of dry leaves and small plant stems, lined with shreds of softer vegetable matter. The nests are built in the forks of bushes, from 3 to 8 feet from the ground, and have the entrance on the lower side. One of the men found a nest situated as described and about 3 feet from the ground. He saw the owner peering out of a hole near the lower side, but as he approached the head vanished, and the entrance was suddenly closed by the opossum drawing some of the nest material across it. The nest was quickly thrust into a game bag, and when examined was found to contain a female opossum and a number of young clinging to her fur with their feet and tails twined closely about hers. The weight of the young was so great that the parent could only walk very slowly. Oryzomys uelsoni Merriam. Nelson's Rice Rat. Oryzomys nelsoni Merriam. Proc. Biol. Soc. Washington, XII, p. 15, Jan. 27, 1898. Type from Maria Madre Island. This rice rat is probably a rare species, as only a few specimens were secured after much trapping. They were found only in damp places near springs about the summit of Maria Madre, about 1,800 feet above sea level. This seemed the most suitable location for them on account of the juicy herbaceous vegetation mingled with the undergrowth. Peromyscus madrensis Merriam. Tres Marias Mouse. Peromyscus madrensis Merriam. Proc. Biol. Soc. Washington, XII, p. 16, Jan. 27, 1898. Type from Maria Madre Island. This is the mostly widely distributed and probably the most numer- ous rodent. Specimens were taken on the three large islands, but its occurrence on San Juanito, where land crabs are very numerous, is doubtful. They were generally distributed over the forest-grown slopes bordering the shore, above the belt infested by crabs. On Maria Madre they were most common about the wild fig trees near the summit (1,500 to 1,800 feet), where the pigmy opossums were secured. Here their burrows entered the ground under logs or projecting roots, but elsewhere these mice were found living beneath rocks and small ledges. They are apparently restricted to the forest, and while nowhere so abundant as were the rabbits iu one place near the north end of Maria Madre, yet they were much more generally distributed. Mus rattus Linn. Black Rat. These rats were found in small numbers about the houses and dis- tributed over the forested parts of Maria Madre and, as on the main- land of western Mexico, we found only the gray form. Lepus graysoni Allen. Tres Marias Cottontail. Lepus graysoni Allen, Mon. N. Am. Rodentia, pp. 347-348, 1877. Type from Tres Marias Islands (undoubtedly from Maria Madre). The cottontail is abundant in some places on San Juauito, Maria Madre and Maria Magdalena, and was reported to occur on Maria MAMMALS OF THE TRES MARIAS ISLANDS. 17 Cleofa. They were very numerous about a deserted rancli on the north side of Maria Magdalena, but were rather scarce elsewhere on that island. We found them extraordinarily abundant and surprisingly tame about old fields on an abandoned ranch at the northern end of Maria Madre. Some were killed with stones near camp, and it would have been easy to kill over a hundred in a morning. They would sit in their forms among the bushes while one peered at them from a dis- tance of a few feet, and when driven out into an open space they often sat quietly while the camera was brought up and focussed within a short distance. The old fields at this ranch had been long abandoned and were covered with a scattered growth of bushes, which seemed more suitable for the rabbits than the forested areas, where they occurred much more sparingly. The cottontails frequented the wood roads leading from the shore up over tbe forested slopes, and after 3 o'clock in the afternoon could be found sitting quietly in little open places in the undergrowth waiting for the nearer approach of sunset before coming out into the roads. The skin of these rabbits was surprisingly delicate, and it was difficult to skin them without tearing it in many places. It was found almost impossible to carry a specimen by the hind legs even a short distance without having the skin tear and slip where it had been grasped by the hand. It is strange that tbe rabbits are not more abundant on the islands, considering the fact that the raccoon is the only predatory mammal, and that the few red-tailed hawks and caracaras are the only birds that prey upon them. Procyon lotor insularis Merriam. Tres Marias Raccoon. Procyon lotor insularis Merriarn. Proc. Biol. Soc. Washington, XII, p. 17, Jan- uary 27, 1898. The raccoon was rather common on Maria Madre and Maria Magda- lena, but no signs of them were seen on Maria Cleofa, where, however, they may occur. In May they were feeding on wild figs and other fruits and on the crabs, which were very abundant near the shore. Every morning freshly made raccoon tracks were seen in trails leading from the seashore to higher parts of the islands, but the animals usually passed our traps without paying the slightest attention to the bait. They were semi-diurnal in habits and several were seen in the woods in broad daylight. One afternoon one was seen crossing the bed of a dry wash near the northern end of Maria Madre, and instead of trying to escape through the woods it climbed a wild-fig tree on the bank and stood looking down from a horizontal branch until shot. Zalophus californianus (Lesson). Sea Lion. A large seal or sea lion, called ' lobo marine' or sea wolf by the Mexi- cans, was reported to occur at several places on the rocky shores of Maria Magdalena and Maria Cleofa. We first heard of them before 13950— No. 18 NORTH AMERICAN FAUNA. leaving San Bias and again upon reaching the islands. It was evident that the sea lions had been hunted for sport by previous visitors until they had become comparatively scarce and are now in a fair way to become extinct. After learning the location of the most frequented places on both islands, we visited them under the guidance of a tortoise- shell hunter who was very familiar with the shore, but we saw only a single sea lion. It was on a rocky islet off the shore of Maria Cleofa, and took to the water and disappeared before we could get a shot. Our guide said that sometimes the sea lions leave the islands for a few days, and this may account for the failure to find them about their usual haunts. The consensus of opinion among the residents of Maria Madre was that these animals are now very scarce. Formerly they were found in many places; but at present a rocky point on the north- west side and a jutting reef on the south side of Maria Magdalena and some islets west of Maria Cleofa are the only landing places used. It is possible that the Guadalupe Island fur seal (Arctocepltalus townsendi Merriam) may also occur at times about the islands. Rhogeessa parvula H. Allen. Tres Marias Rhogei-ssa. Ehogeessa parvula H. Allen. Proc. Acad. Nat. Sci. Phila., 1866, p. 285. Type from the Tres Marias. These little bats were rather common on Maria Madre, where they live in the forest and fly at dusk along the trails and about small open places. At times they appear in such situations in broad day. Two were killed while flying up and down a trail in the brilliant sunshine in the middle of the forenoon, and I saw one hawking for insects among the tree tops along a trail two hours before sunset. As a rule, however, they only come out when it is too dark for one to see more than an indistinct form as they flit about among the trees. A few were also seen on Maria Magdalena. Myotis nigricans (Maximilian). Maximilian's Black Bat. According to Mr. Oldfield Thomas, a specimen of this bat was taken on the Tres Marias by Mr. Forrer.1 We took none, and they probably occur on the island only as stragglers. Otopterus niexicanus (Saussure). Big-eared Bat. A colony of over a hundred big-eared bats was living in an old ware- house at the settlement on Maria Madre, and others were found in sev- eral caves situated in various parts of the island. The warehouse where these bats were found had a large open window and wide cracks, so that it was quite light inside, yet they were found hanging from the ceiling and roof, in plain view, and evidently had lived there a long time. The specimens were mostly females heavy with young. Glossophaga mutica Merriam. Tres Marias Glossophaga. Glossopliaga mutica Merriam. Proc. Biol. Soc. Washington, XII, pp. 18-19, Jan- uary 27, 1898. Type from Maria Madre Island. This was by far the most numerous bat on Maria Madre, where it was found in every cave sufficiently deep to be dark. One cave was among 1 Biologia Central!- Americana, Mammalia, 206, 1881 (under Fespertilio nigricans). MAMMALS OF THE TRES MARIAS ISLANDS. 19 some huge projecting rocks lying at the water's edge, near the settle- ment. Many of the females collected contained large embryos. These bats were feeding on the fruit of the wild fig. As surmised by Dr. Merriam,1 the record of Chceronycteris mexicana from these islands, given by Mr. Thomas in the Biologia, proves to be referable to the present species. In reply to a letter of inquiry, Mr. Thomas states that he discovered the mistake in identification too late to correct it in the Biologia, and agrees with Dr. Merriam in referring his specimen to G. mutica. Lasiurus borealis mexicamis (Saussure). Mexican Red Bat. Forrer added this species to the fauna of the Tres Marias as recorded by Mr. Thomas.3 We did not see any red bats, and I doubt their being found on the islands except as stragglers from the mainland. Bats are such wide ranging animals it is to be expected that several addi- tional species will eventually be found to occur on the islands. ? Phocaena commimis (Lesson). Common Porpoise. Porpoises supposed to belong to this species were common around the shores of the Tres Marias and also in bays and mouths of streams or lagoons along the coast of the mainland. They were always seen in the belt of shallow discolored water within a short distance of the shore. As soon as blue water, with a depth of over 40 fathoms, was reached, the other porpoise (Prodelphinus longirostris) was encoun- tered. The common porpoise was seen in schools of 10 to 30 or 40 indi- viduals swimming in loose order. At Maria Madre they came into the shallow bay in front of the settlement in the early morning and followed close along shore. Prodelphinus longirostris (Gray). Long-nosed Porpoise. In the blue water between the mainland and the islands these por- poises were very abundant in schools of from 100 to 200 individuals. They are much slenderer and more graceful animals than the preceding species. While swimming about their feeding places at sea they were accompanied by swarms of terns, gaunets, and shearwaters. On one occasion, while crossing to the islands, a school of about 200 porpoises came directly toward us and passed under and on all sides of the boat. While they were passing, the water was broken into foam on every hand by their glistening black bodies, and overhead swarmed a shriek- ing crowd of sea birds. Mr. Goldman made a fortunate rifle shot and killed two of them, but one sank before it could be harpooned. 'Proc. Biol. Soc. Washington, XII, pp. 13-19, footnote, January, 1898. 8 Biologia Ceutrali-Americaua, Mamm., p. 205, footnote. BIRDS OF THE TRES MARIAS ISLANDS. By E. W. NELSON. The present paper is based mainly upon the birds found on the Tres Marias, but for the sake of completeness the results of our work on Isabel Island have also been introduced.1 The situation of Isabel Island between the mainland and the Tres Marias renders its bird life of peculiar interest in the present connection. Mr. Xantus sent speci- mens of birds to the National Museum labeled 'Tres Marias, 1861,' but only one of these can be an authentic island species, and it seems almost certain that Xantus did not visit the islands. Colonel Grayson's notes on his three visits to the group and his trip to Isabel Island were published by George N. Lawrence in the 'Proceedings' and 'Memoirs of the Boston Society of Natural History,' while the descriptions of new birds in his collections appeared in various publications and are mentioned in the bibliography (see pp. 93-94). Gray sou constantly refers to the 'Tres Marias Islands,' but the internal evidence of his writings, in addition to the information given me by the inhabitants, indicates that all of his work was done on Maria Madre. Mr. A. Forrer visited Maria Madre in 1881, but the publication of Vireo flavoviridis forreri by Von Madarasz and a few notes in the 'Biologia Centrali- Americana' and in some of the British Museum Catalogues are all we know of his work there. As already stated in the general introduction, our work was done on Isabel Island on April 22 and 23, on Maria Madrc from May 2 to 25, and six days were spent working about Maria Magdalena and Maria Cleofa. It is quite certain that the bird fauna of Maria Madre is now fairly well known, and it will be advisable for anyone visiting this group in the future to give attention chiefly to the two smaller islands. It is certain that a large proportion of the birds found on Maria Madre occur also on Maria Magdalena, but some of the species living in the dense forest at higher altitudes on these islands probably do not occur in the more scanty forest of Maria Cleofa. At present 83 species and subspecies of birds are known from the Tres Marias, and further observations will, no doubt, add to the list various stragglers from the mainland. The bird fauna may be grouped under the following headings : Resident land birds, 3G species or sub- 1 The notes iu the following pages refer to Isabel Island only when BO stated. 22 NORTH AMERICAN FAUNA. species. Visitant land birds, 26 species or subspecies. Eesident water fowl, 13 species. Visitant water fowl, 8 species. Of the 36 resident species or subspecies of land birds all but 5 were observed by Colonel Grayson. These exceptions are: Melopelia leucop- tera, Tyrannus melancliolicus couchi, Ornithion imberbe, Vireo flavoviridis forreri, and Thryothorus lawrencii magdalence. Twenty-four of the 36 resident land birds are specifically or subspe- cifically distinct from their mainland representatives. Of this number 12 were described from Grayson's collections, 1 from Ferrer's, and 11 from our own. A study of our collections from the islands, and near San Bias on the mainland, brings out the interesting fact that several species from the latter district show a decided approach to their island representatives. This is very marked in Compsothlypis which is very nearly the same at San Bias as on the islands. The Polyborus and Platypsaris from that locality seem to be intermediate between the island races and the birds of the mainland. Specimens of Thryothorus felix from the same part of the coast are much nearer T. lawrencii than they are to typical T. felix. Among the 24 species or subspecies of land birds peculiar to the islands 15 are larger than their relatives of the nearest mainland. These are Columbaf. madrensis, Leptotila capitalis, Psittacula insularis, Dryobates s. graysoni, Nyctidromus a. insularis, Amazilia graysoni, Icterus graysoni, Cardinalis c. maricc, Piranga b. flammea, Vireo /. forreri, Vireo h. sordidus, Compsothlypis insularis, Granatellusfrancesca;, Thryothorus lawrencii and Merula graysoni. Six of the island birds average smaller than their mainland repre- sentatives. These are Polyborus c. pallidus, lache lawrencei, Platyp- saris a. insularis, Myiopagis p. minimus, Melanotis c. longirostris, and Trogon a. goldmani. The two first named are generally smaller, but Platypsaris a. insular is has a longer tarsus, Myiopagis p. minimus a longer bill and tarsus, Melanotis c. longirostris a longer bill, and Trogon a. goldmani a longer bill and tarsus. Although Compsothlypis insularis also occurs in a limited area along the coast, I have considered it as a typical island species. The difference in size between island birds and their mainland repre- sentatives varies greatly, being slight in some and very well marked in others. Nyctidromus a. insularis is a larger bird than albicollis proper, but has a shorter bill and tarsus. Among the birds peculiar to the islands Thryothorus laicrencii magdalence and Myadestes o. insu- laris are almost the only ones which do not show more or less well- defined differences in size from their nearest mainland relative; a series of the first named, however, may show that it also differs. One of the most puzzling features of the fauna of these islands is the absence of various land birds found on the adjacent mainland. Although the physical conditions appear so much like those of the mainland, yet some change must have occurred to upset nature's fine balance and render these isolated areas unsuitable for many species. BIRDS OF THE TRES MARIAS ISLANDS. 23 The death by starvation of the Louisiana Tanagers on Maria Madre Island (p. 52) is an example of the manner in which the island fauna may be maintained in its present state. As the climatic conditions on the islands and on the mainland are very similar and the vegetation nearly alike, this paucity of species presents one of the curious prob- lems of distribution. It would be hard to find an equal area of similar country on the mainland, near San Bias, where so few species of land birds could be found. The only reasonable explanation seems to be the scarcity of water and the long, dry season, which combine to reduce the food sup- ply and perhaps reuder the country unsuited to some species. It was very surprising to find a total absence on the islands of such com- mon and widely spread mainland genera as Conurus, Momotus, Piaya, Campephilus, Melanerpes, Myiozetetes, Cissolopha, Cyanospiza, Pipilo, Pyrgisoma, Saltator, and others. The absence of Pipilo is especially unexpected, for this genus is represented on Socorro Island, which lies very much farther at sea off the same part of the coast. ANNOTATED LIST OF SPECIES. ? Brachyrhamphus brevirostris (Vigors.) Short-billed Murrelet. Brachyrhamphus brevirostris was described from San Bias and B. hypoleucus from Cape St. Lucas. Colonel Graysou mentions having seen 'guillemots' at Isabel Island (Mem. Boston Soc. Nat. Hist., II, p. 318, 1874) and off the Tres Marias group (Proc. Boston Soc. Nat. Hist., XIV, p. 288, 1871). This led me to anticipate finding at least one of the species there, and it is with some disappointment that I have to record our failure to see either species about the islands, although I watched for them constantly. From this experience I am inclined to think that they breed only along the coast of Lower California, and visit these islands sporadically. Larus argentatus smithsonianus Cones. American Herring Gull. A single immature specimen was taken on San Juanito Island May 22, and a few others were seen. These birds were flying back and forth along a strip of beach where a large colony of blue footed gannets were breeding, and the gulls probably had an eye on the nesting ground for the purpose of capturing any unprotected eggs. They were noted singly a few times along the shores of the Tres Marias and at Isabel Island. No fully adult individuals were seen. Larus heermanni Cassin. Heermann's Gull. * On April 23 a fine adult bird of this species was shot on the shore of Isabel Island. In company with its mate it had harried a blue-footed gannet into disgorging a number of small fish upon a rock at the edge of the water, and was picking up the spoils by a series of little down- ward swoops and hoverings. The gannet had shuffled into the water and was making off, with backward glances at its tormentor, when I drew near. These gulls are bold and noisy aggressors when they wish 24 NORTH AMERICAN FAUNA. to take advantage of the gannets, and about the breeding places of the latter they feed largely at the public expense. But few of them were seen about the islands — two or three pairs at Isabel and half a dozen pairs about the Tres Marias. A nest, which had been occupied earlier in the season, was seen on the ledge of a rocky islet off the shore of Maria Oleofa May 30, and full grown young of the year were also seen on the rocks. Sterna maxima Boddaert. Royal Tern. None were seen at Isabel Island, although they were not uncommon during April along the mainland coast. During May they were seen in small parties about the shores of all the Tres Marias group, where they probably breed in very limited numbers. The only specimen saved was taken May 31 from a flock of six which was coasting along the beach at Maria Cleofa. Sterna elegans Gambel. Elegant Tern. Sterna galericulata Lawr., Mem. Boston Soc. Nat. Hist., II, p. 317 (1874). Colonel Graysoii found these terns on Isabel Island, but none were seen by us. Sterna fuliginosa crissalis (Lawr.). Pacific Sooty Tern. Haliplana fuliginosa var. crissalis Lawr. (ex Baird MS.), Proc. Boston Soc. Nat. Hist., XIV, pp. 285, 301, June, 1871; Mem. Boston Soc. Nat. Hist., II, p. 318, 1874. These handsome terns are common about Isabel Island where Colo- nel Grayson found them breeding. My observations from the last of April to the first of June led me to believe that at this season Isabel Island is their central roosting point. During the week we were cruis- ing about Isabel and the Tres Marias islands many flocks were seen. From about noon until the middle of the afternoon or later the flocks were generally flying directly toward Isabel at an altitude of from 50 to 200 yards above the water. This was noted also near the islands, while we were crossing the straits between the Tres Marias, and off the mainland near San Bias. Many of the birds were perched along the top of an inaccessible rock just off Isabel, and were also seen alighting on the cliffs of the northern and northeastern side of the island, but the boat was too unwieldy for us to venture near enough to closely examine these haunts. The birds have a peculiar shrill cry which they often utter while feeding and when flying about at night. The night before we landed on Isabel Island it was necessary to anchor about midway between the island and the shore. The wind blew strongly in the afternoon but fell at sunset, a dead calm ensued, and heavy clouds overspread the sky. During the day only a few sooty terns had been seen, but from about 9 p. m. until near daybreak they were evidently much more numerous, for their cries were heard at short intervals. Several times the notes were uttered directly overhead and the birds seemed to be scarcely higher than the top of the mastT where they apparently paused and hovered while they examined the BIRDS OF THE TRES MA.RIAS ISLANDS. 25 boat with great curiosity. As they were heard every night while we were at sea, it is evident that they were both diurnal and nocturnal in habits. They feed well out at sea, and were not found anywhere along shore, except when they came to their roosting place on Isabel Island. There were no signs of their roosting about the Tres Marias, although they may roost on some of the outlying rocky islets. Grayson found them in small numbers farther west, about the Eevillagigedo Islands. During our trip to the Tres Marias many schools of large fish were encountered swimming close to the surface and constantly breaking, often with such force and rapidity that tbe water boiled and foamed over considerable areas. These schools of fish were commonly acconi. pauied by flocks of sooty terns and gannets, which appeared to be animated by the wildest excitement. The terns hovered over the foaming sea, uttering shrill cries and darting down into the water, evi- dently after food; and in the midst of the turmoil the blue-footed gau- nets swam about, beating the water with their wings and adding to the noise made by the terns and leaping fish. While on Maria Madre I saw a flock of terns some distance oft' shore, and taking a canoe, managed to get out to them, and directly in the course of tbe school of fish they were accompanying. Letting the boat drift, I stood up and watched the swarm go by. Thousands of large fish and hundreds of terns and ganuets passed the boat on every side, amid loud cries from the terns, a rushing sound from the fish and ganuets, and a bewildering complexity of motion in sea and air that was intensely exciting. This novel sight was so interesting that I came near losing the chance to secure some of the birds. These terns were seen also following schools of porpoises oft7 shore — in the latter case accompanied by the wedge-tailed shearwater. In the passage between Maria Magdaleua and Maria Cleofa a flock of sooty terns was seen soaring in wide circles high overhead and finally start- ing off for their roosting place on Isabel Island. The 'variety crissalisj named in manuscript by Professor Baird and published by Mr. Lawrence, was characterized as "having the under tail coverts tinged with ashy, instead of being pure white." A series of 17 specimens from the west coast of Mexico, and from widely scat- tered islands of the Pacific and Indian oceans, agree in having the posterior part of flanks, under wing coverts, and entire crissum dis- tinctly ashy, not a single individual being white on these parts, as is commonly the case with birds from the Atlantic and Gulf coasts of North America. Unfortunately the series of Atlantic birds at hand is very small, but there is little doubt that crissalis is a valid subspe- cies. Birds from the west coast of Mexico, the Galapagos Islands, and Hawaii agree in having an average shorter bill and tail than those from elsewhere. Specimens from the Indian Ocean have even a longer bill and tail than thos* from the Atlantic, but are ashy below. 26 NORTH AMERICAN FAUNA. like those from western Mexico. Specimens from Ascension Island, off the west coast of Africa, also have a light ashy shade on the lower tail coverts. The following average measurements show the sizes of these birds from various parts of their range : Table of measurements of Sterna fuliginosus and Sterna f. crissalis.* Name. Locality. Num- ber of speci- mens. Wing. Tail. Cul- men. Tarsus. Sterna fuliginosus . . East coast of North America and 7 288.1 151 42.8 2X5 west coast of Africa. Sterna fuliginosus West coast of Mexico, Hawaii, and 10 288.6 143. 5 41.8 23.6 crissalis. Galapagos Islands. Sterna fuligiuosns Krusen stern Islands (west of Hawaii) . 2 292.5 203.5 39.5 24.2 crissalis. Sterna fuliginosus Glorioso Island (Indian Ocean) 3 292.6 192 43.3 23.8 crissalis. * All measurements are in millimeters. In the foregoing measurements the length of the tail is unreliable, owing to its variability, on account of wear and other causes. Anous stolidus ridgwayi Anthony. Pacific Noddy Tern. Anous stolidus Lawr., Mem. Boston Soc. Nat. Hist., II, p. 318, 1874. Anous stolidus ridgwayi Anthony, Auk, XII, p. 36, 1898. Common the last of April on Isabel Island, and a few seen off the Tres Marias during May. Between San Bias and the islands a number of these birds were seen. We usually saw one or two indi- viduals at a time, and did not find them in flocks anywhere except when congregated on the rocks at their roosting places. At sea they usually flew close along the surface of the waves with long, graceful wing strokes. From their dark color and habit of keeping close to the water they were several times mistaken for black petrels. They were found in considerable numbers on the ragged faces of cliffs and rocks along the northeastern point of Isabel Island, and were very unsuspicious, permitting us to approach quite near in the boat. While perched on the black lava cliffs, their dark color blended so closely with the background that it was very difficult to distinguish them, even when within fair gunshot. The day we left the island we visited their resting place and fired a dozen or more shots while they were on the rocks or flying about, but the noise of the reports did not seem to give them much alarm. They would circle out a short dis- tance, and, after hovering for a few moments over their killed or wounded companions floating in the water, would return to the same part of the cliff from which they had just been startled. They were not heard to utter any notes, and the silence with which they would suddenly appear out of the cliff and then return and vanish again in its gloomy face produced an uncanny effect. BIRDS OF THE TRES MARIAS ISLANDS. 27 Colonel Grayson found them nesting on the north end of Isabel Island in April, 1869, and states that they were breeding in communi- ties on shelving rocks beneath overhanging cliffs. The nests were placed close together, but were inaccessible. A single egg was pro- cured, however, which was white, with scattering brownish blotches, most numerous about the larger end. This is undoubtedly the tern which Colonel Grayson reports as replacing the sooty tern on the Revillagigedo Islands, and which he described as being black, wit'a a hoary forehead. Puffinus cuneatus Salvin. Wedge-tailed Shearwater. Puffinus cuneatus Salvin, Ibis, 5th ser., VI, p. 353, July, 1888. Fttffinus knudseni Stejn., Proc. U. S. Nat. Mus., XI, p. 93, Nov. 8, 1888. During our trip to and from the islands we saw 100 or 200 wedge- tailed shearwaters. They were usually seen singly skimming along over the sea, at an elevation of a few yards, making widely sweeping circuits and pausing occasionally to pick up bits of food. When about midway between Isabel Island and the Tres Marias we encountered several schools of small porpoises of 150 or more individuals, which traveled in close array, frequently gamboling about and playfully leaping high in the air. A swarm of sooty terns followed the porpoises, and twice when they passed near us I saw considerable numbers of these shear- waters among the terns. Judging from the numbers, they must be rather common in these waters, but none were seen near the islands. This species was first described from specimens taken on the Kru- senstern Islands, in the Marshall Group, and Mr. A. W. Anthony made the first record of its occurrence on the American side of the Pacific, at the Revillagigedo Islands, during the summer of 1897 (Auk, XV, Jan., 1898, p. 39). As it is a species new 1o North America, a detailed description of our specimen is given, in order to facilitate identification in case other examples should be taken. Description.— No. 156678, U. S. Nat. Mus., Dept, of Agriculture Coll. Ad. $ , off Maria Madre Island, May 2, 1897. Collected by E. W. Nel- son and E. A. Goldman. Top and sides of head and neck grayish-brown ; forehead, lores, and space from latter area back beneath eyes and along sides of neck paler or more ashy, thus edging the darker area of the crown and upper neck with lighter. Back, including rump and upper tail coverts, mainly dark brown, but mixed with numerous feathers of a decidedly grayish, almost ashy, shade. These latter feathers undoubtedly indicate the color of dorsal surface in fresh-plumaged birds. Wings and tail blackish-brown. Entire lower surface of body white, shaded with dingy ashy, darkest ou sides and palest along median portion. Under tail coverts mixed dark brown and grayish-brown. Border of the wing along under side brown or grayish- brown; under coverts white with a little necking of pale gray on some feathers. In the flesh this specimen had a horn- blue bill with flesh-colored feet and tarsi. 28 NORTH AMERICAN FAUNA. Dimensions.— Wing, 293; tail, 135 (length of middle pair of tail feath- ers beyond lateral pair, 47); culmen, 42; tarsus, 50; middle toe, with claw, 59. Habitat. — The range of this species is now known to extend across the middle North Pacific from Japan to the west coast of Mexico. Oceanodroma melania (Bonap.). Black Petrel. Common between Isabel and the Tres Marias. Black petrels were by far the most numerous of the petrels seen, and outnumbered all the others two to one. Three, and possibly four, other species were seen on the way to and from the islands, but this was the only one secured. They circled about in all directions, sometimes coming very near, but nothing peculiar in their habits was observed. They were quick to see little fragments of fat thrown overboard while we were skinning other waterfowl, and when the morsels were small enough ate them greedily. Phaethon aethereus Linn. Red-billed Tropic Bird. Tropic birds are readily distinguished on the wing by their graceful ternlike flight and long filamentous tail feathers. Many of them breed on Isabel Island and in suitable places on rocky islets near San Bias and about the Tres Marias. The last of April fresh eggs and nearly grown young were found on Isabel, and by the last of May the young on the Tres Marias had taken wing and few were to be seen, although we were told by the tortoise shell hunters that many breed there earlier in the season. Soon after landing on Isabel, a tropic bird was found sitting on its sol- itary egg at the end of a little hole in the rock close to the beach. The hole was only about 15 or 18 inches across and about 3 feet deep, so that there was no difficulty in taking the bird by hand after a little maneu- vering to avoid its sharp beak. During a stay of about twenty-four hours on this island at least 20 nests containing eggs or young were examined. A single egg is laid directly on the rough rock or loose dirt forming the floor of the nesting site, which is always located under the shelter of over arching rock, but varies greatly in situation. The inner ends of holes in cliffs facing the sea were favorite places, but as the number of such situations was limited, the birds were forced to utilize small caves and even rock shelters. In one locality five or six nests were placed on loose earth at the bottom of rock shelters so situ- ated that 1 could walk directly up to them and pick up the birds. Whenever a nest was approached the parent screamed and fought viciously, ruffled its feathers and looked very fierce, but made no attempt to escape. They protested with beak and voice when pushed about, but as soon as I went away a few yards they would shuffle back to resume their former position over the egg. The young, even when quite small, were equally fierce in resenting any intrusion. One nest was found on the beach under the edge of some great rocks that had fallen from the adjacent cliff'. It was only 5 or C feet above high tide BIRDS OF THE TRES MARIAS ISLANDS. 29 and would have been overlooked but for the angry cries of the old bird when she heard me walking over the roof of her habitation. "At, sunrise the old birds were found sitting side by side at the mouths of their nesting places waiting to enjoy the first rays of sunlight. Half an hour later one of each pair started out to sea while the other resumed its place on the nest. When disturbed on the nest their cries are very shrill and strident, consisting of a series of short, harsh, clicking or rattling sounds something like the noise of an old-fashioned watch- man's rattle. The young are covered at first with fluffy white down. Before they are one-third grown the first plumage begins to appear, and is very similar to that of the adults, except that the black barring on the back is broader. Sula websteri Rothschild. Webster's Booby. Sula bassana Grayson, Proc. Boston Soc. Nat. Hist., p. 302, 1871. Sula websteri Rothschild, Bull. Brit. Orn. Club, VII, No. LIV, p. LII, 1898. This booby is no doubt the Sula bassana reported from Isabel Island by Grayson but not seen by us. Mr. Anthony found it the most abun- dant species breeding on the Revillagigedo Islands during the summer of 1898. Sula brewsteri Goss. Brewster's Booby. Sula brewsteri Goss, Auk V, p. 242, 1888. Brewster's boobies were very numerous on a small hill at one side of the little bay where we lauded on Isabel Island April 22, but there were no signs of their breeding. They came in from sea during the first half of the afternoon and sat about on rocky parts of the shore until nightfall. Scattered individuals were also seen about the ledges and tops of the cliffs facing the sea. T.he following morning at day- break they were congregated on the little hill already mentioned which is probably their regular roosting place. About half an hour after sunrise they began to start out to sea singly and by twos and threes until all were off on the day's fishing expedition. A few were seen about the rocks just oft' San Bias, and were said to breed on the large rock (Piedra Blanca) midway between Isabel and San Bias. Only a few of these boobies were seen about the Tres Marias until an islet was visited off the northwest shore of Maria Cleofa. This islet rises from 150 to 200 feet above the sea, with cliffs on all sides. The sum- mit is mainly rolling, with an elevated, sloping bench on one end At this time, May 30, many thousands of boobies were breeding on the bare top of this rock. The eggs were laid directly on the surface, with no sign of a nest. The sun was intensely hot and heated the rocks so that they were uncomfortably warm to the touch. The birds did not sit upon the eggs during the hottest hours, but while standing to avoid contact with the heated rocks kept in such position that the eggs or young were shaded from the sun, and thus had their vitality preserved. While trying to secure photographs of this breeding ground a few of the old birds flew away and it was surprising to see how quickly the 30 NORTH AMERICAN FAUNA. newly hatched young succumbed to the heat when the parents left them exposed to the rays of the sun. The nests were spaced at intervals of 4 or 5 feet, so that the old birds were safely out of reach of one another. Although so gregarious in their breeding habits, they appeared to have but little regard for one another. It was amusing to see the sav- age way in which the nest owners assisted intruders of their own kind out of their territory. While we were walking among them some of the birds would often waddle off to one side, and in so doing neces- sarily trespassed on their neighbors. The latter at once raised a hoarse shrieking and set upon the outsiders with wicked thrusts of their beaks, which continued until the victims took wiDg and escaped. We were also subjects of this proprietary rage, and had our legs nipped every now and then, despite all efforts to walk circumspectly. Our progress over the breeding ground was accompanied by a wave of hoarse, nasal cries that sometimes became almost deafening. Many of the birds were valiant upholders of their rights and sturdily refused to leave their nests, which they defended vigorously, all the time utter- ing loud cries of rage. These birds show very little individual variation in color. As the species is not well known the following descriptions are appended from specimens taken on Isabel and Maria Cleofa. Adult male. — Nearly entire head white, shading gradually on poste- rior portion into drab of neck and then insensibly into dark, sooty brown of back. On lower side of neck the drab becomes darkest at poste- rior border, where it ends abruptly against the pure white of lower parts. Bill light horn color; gular pouch in life livid blue; feet green- ish yellow — the latter varying in intensity. Male in immature plumage. — Dorsal surface uniform dark brown, slightly paler than back of adult; entire lower surface still paler and more dingy brown. Feathers over much of body, especially about head, neck, and lower parts, narrowly edged with grayish brown, giving a faint wavy barring. Bill bluish horn color, with darker shade of same about base and on gular pouch ; feet and tarsi dull fleshy yellow; iris greenish gray. Adult female. — Head, neck all around, and back sooty brown ; ven- tral surface below neck white. Bill light horn color; a spot of leaden bluish on lores ; base of bill, gular pouch, feet, and tarsi grayish yel- low; iris pale grayish. Average measurements of these birds from Isabel Island are as fol- lows: $ (5 specimens), wing 384.4; tail 189.6; culmen93.6; tarsus 45.4; $ (5 specimens), wing 416.6; tail 192.8; culmen 96.6; tarsus 48.8. Nestlings a few days old are covered with fluffy white down. A male bird of the previous year, which still retained the immature plumage, was taken at Isabel on April 23, and several others were seen. BIRDS OF THE TRES MARIAS ISLANDS. 31 Sula nebouxii MilnerEdwards. Blue- footed Booby. Sula pisoator Grayson, Proc. Boston Soc. Nat. Hist., XIV, p. 302, 1871; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 316, 1874. Sula nebouxii Milne-Edwards, Ann. Sci. Nat., Paris (Zool.), 66me s6r., XIII, Art. 4, p. 37, pi. 14, 1882 (Chile). Sula gossi Goss (ex Ridgway MS.), Auk, V, p. 241, July, 1888 (Gulf of Calif.). Sula nebouxii is the most abundant species of booby occurring on Isabel and the Tres Marias. On April 22 they were breeding abun- dantly on the beaches and on a low flat area that covers a part of the former island. They were common on the grassy beach at the landing and thence back among the scrubby trees and bushes which form a scanty growth over the flat. They were most numerous on the open beach a little above high-water mark, but dozens of them were seen with their eggs farther back among the bushes. Like the preceding species, they fought and screamed savagely when approached. The males usually flew away, but the females remained to give battle over the nests, which were mere hollows in the earth, sand, or gravel. Not a single young one was seen in the hundreds of nests on Isabel. The sun was excessively hot the morning of our arrival, and while the men were landing the outfit, ropes were fastened between the tops of some scrubby trees close to the beach and a piece of canvas spread for an awning, under which the baggage was placed. An old booby had her eggs in the sand within 3 feet of the edge of the sheltered area and stood her ground unflinchingly while the men were at work, keeping a wary eye on their movements and making vicious dabs whenever a leg came incautiously within reach. Having arranged camp, I went out exploring for an hour or so and returned with various specimens, including the egg of a tropic bird, and found that one of the boatmen had driven off the booby and thrown away her eggs. Wish- ing to test the bird's discrimination, I placed the reddish-brown egg of the tropic bird in the hollow where the two greenish- white eggs of the booby had been, and sitting under the awning began to prepare specimens. In the course of half an hour the owner of the despoiled nest returned and alighted 10 or 15 steps away near another deserted nest, gave a look at the eggs in it, walked to still another, looked at it, and then proceeded directly to her own nest and stopped. She looked about and then down at the nest. The presence of the single reddish- colored egg appeared to surprise her; she looked at it with one eye and then with the other as if in doubt. An instant later the feathers on her head and nape ruffled up and with a loud squawk of rage she suddenly dashed her beak again and again into the strange egg, breaking it to fragments in a moment. As soon as the egg was demolished she took wing and disappeared out to sea. There was no intention to sacrifice the tropic bird's egg in this experiment, so the « booby carried off the honors. About 10 o'clock the following night a visit was paid to the nesting boobies. The night was calm, and taking a lighted candle I walked 32 NORTH AMERICAN FAUNA. out a short distance to an opening in the bushes where there were twenty or thirty nests. The females were found on their eggs with the males standing close beside them. When the strange visitor appe'ared in their midst the birds set up a continuous series of hoarse cries and, like so many moths, seemed to become fascinated by the light. They started up on all sides, and trooping within the circle of bright light, began to run around me in a ring about 20 feet in diameter. They ran in single file from right to left and presented a most ludicrous sight. Occasionally one fell on its breast, whereupon the others scrambled over the fallen bird until it regained its feet and rejoined the proces- sion. One of the number was suddenly possessed with a desire to run around one of my legs, and, although seized by the head several times and tossed out among its companions, persisted in returning to the same place and continuing its gyrations. The next morning at day- break the birds were seen standing in pairs by their eggs and remained in this position until about sunrise, or a little after, when all of the males went out to sea — usually in little parties of two to five or six. They returned between 1 and 3 o'clock in the afternoon, and a number of them flew directly to their mates and disgorged numerous small fishes which the females ate greedily. These observations seemed to show that the females did the incubating and the males provided the food. As the neighboring waters do not abound in small fishes, the boobies have to go in many cases from 10 to 30 miles to obtain their daily supply. During a visit to San Juauito Island, the latter part of May, many blue-footed boobies were found breeding on sandy beaches at the south end of the island; many of the young were hatched and some were more than half grown. Like the young of the tropic birds, the young boobies uttered angry cries and fought savagely when approached. This species is found in Chile, on the Galapagos Islands, and north to the island of San Pedro Martir in the Gulf of 'California. The type of Sula nebouxii was obtained on the coast of Chile, and the type of S. gossi came from San Pedro Martir. The sexes are alike in color, but when standing together the males may be readily distinguished by their smaller size and slenderer form. In life the bill is leaden horn color, with its base and the gular pouch leaden blue; the feet are bright blue. The downy young are pure white. Four specimens measure as follows: $ (1 specimen), wing, 410; tail, 219; culinen, 108; tarsus, 51. ? (average of 3 specimens), wing, 438; tail, 219; culiuen, 109.6; tarsus, 56. Phalacrocorax sp. Cormorant. Two or three cormorants seen at a distance were the only ones noted during the trip. None were seen near Isabel nor on the rocks near San Bias. Pelecanus californicus Ridgway. California Brown Pelican. A few pairs of brown pelicans were breeding on Isabel Island the last of April. The nests were made of sticks and placed in the dense BIRDS OF THE TRES MARIAS ISLANDS. 33 tops of the scrubby trees growing on the rocky inner slopes of the island. They were found about all of the Tres Marias, but usually occurred singly and were nowhere common. Two or three were seen fishing in the breakers alongshore at the north end of Maria Madre. Fregata aquila (Linn.)- Man-o'-war Bird. Hundreds of inan-o'-war birds were breeding on Isabel in April, and on approaching the island many were seen soaring over the rocky sum- mit. The first shot caused hundreds of others to take wing, and in a few moments the air was swarming with them. They soared in con- stantly intersecting circles, until the sky seemed covered with their silhouetted outlines. So few had been seen alongshore near San Bias that it was an interesting and unexpected sight. Upon landing, numer- ous large, oval, and brilliantly red objects were seen in the tops of the dark-green bushes along the slopes. These proved to be the gular pouches of old male man-o'-war birds inflated to the size of a man's head, the brilliant red color of the distended membrane making them very conspicuous objects. It appeared to be a common custom of the birds to sit quietly on the top of a tree for a long time with the pouches thus distended and evidently serving as sexual ornaments. A few birds were seen circling high overhead with their pouches fully inflated, but as a general rule, when soaring, the pouches were closed. The nests were built of sticks and placed in the tops of low trees and stout bushes from near sea level to the summit of the island. Three or four of these platform-like structures were found together in some of the larger bushes. Many of the young were hatched and, when able to stand alone, would do their best with voice and beak to resent our approach. The young are covered with white down until nearly half grown. A few of these birds were also seen about the Tres Marias, and are said to breed on San Juanito, but were not common. Ardea herodias Linn. Great Blue Heron. A few solitary individuals were seen at various times during May along the beaches, and it is probable that a few pairs may breed on the islands. They were also noted by Colonel Grayson. Ardea egretta Gmel. American Egret. The American egret is another species noted by Colonel Graysou, which we did not see. It must occur merely as a straggler from the mainland. Ardea candidissima Gmel. Snowy Heron. Kecorded by Colonel Grayson as a straggler, but not seen by us. Nycticorax violaceus (Linn). Yellow-crowned Night Heron. A dozen or more night herons were noted and a young bird was shot on a rock close to Isabel Island, where it must have strayed from the mainland. As Colonel Graysou found them on the islands and took 13950— No. 14 3 34 NORTH AMERICAN FAUNA. specimens in immature plumage, it is very probable that they breed there iu small numbers. Tetanus .flavipes (Gmel.). Yellow-legs. A single specimen was shot by Prof. C. L. Herrick on Maria Madre the middle of May. Actitis macularia (Linn.)- Spotted Sandpiper. A few were seen along the shore on all of the islands, where they probably breed. Several small flocks of another sandpiper were seen along shore on the islands; but although considerable energy was expended in their pursuit we failed to secure a specimen. JEgialitis semipalmata Bouap. Semipalmated Plover. Colonel Grayson took a single specimen of this plover. It was not seen by us and must occur only as a straggler. Haematopus palliatus Temminck. American Oyster-catcher. Bather common on the shores of the Tres Marias and Isabel, as well as along the coast of the mainland near San Bias. A series of ten oyster-catchers were secured and have been compared with four speci- mens of Haematopus galapagensis, one of H. frazari, and a number of typical H. palliatus (from the Atlantic coast of the United States and the West Indies). All of the birds from the Tres Marias, Isabel Island, and the adjacent mainland were found to be surprisingly close to typical palliatus. As the Tres Marias are not far south of Lower California, the birds from the islands might naturally be expected to be nearly typical repre- sentatives of //. frazari. In reality about the only sign of gradation toward the latter form is the mixed black and white across the lower border of the black neck area. Some specimens from a single small flock on Maria Cleofa had the line of demarkatiou between the black and white areas on the breast as sharply denned as in palliatus, while others had the mixed black and white areas, as in frazari and galapagensis. Some of the birds have a white spot on the under eyelid, which is absent in others, but otherwise the color is the same as in typical pal- liatus. Birds from the Tres Marias and the mainland coast to the south have an average shorter bill and tarsus than true palliatus, and in this character approximate frazari and galapagensis. Mr. Bid g way has already called attention to the close general similarity existing between the two latter species. The series from the Tres Marias and adjacent coast agree with specimens in the National Museum from various points along the Pacific coast of Mexico, Central, and South America in being very close to typical palliatus, thus showing pretty conclusively that this is the resident bird along the coast and adjacent islands south of Lower California. So far as can be judged from specimens at hand, H. galapagensis is distinct from frazari, although the birds resemble one other more closely BIRDS OF THE TRES MAKIAS ISLANDS. 35 than they do representatives of palliatus from the adjacent mainland. A series of specimens from the southern eiid of Lower California will probably show intergradation between palliatus find Jrazari. The fol- lowing measurements show the comparative sizes of birds from various localities: Measurements of Heematopus palliatus, H.frazari, and H. galapagensis. Name. Locality. Sex. Num- ber of speci- mens. Wing. Tail. Oul- ini-ii. Tarsus. Haematopns palliatus.. Atlantic coast, United ? 3 260 102.3 90 63 States and West Indies. * Hicmutopus palliatus.. T res Marias and Isabel 68, 1871. A single duck hawk was taken by Colonel Grayson, who mentions that it was shot while in close pursuit of a sparrow hawk. We did not see this species during our visit to the islands. Palco sparverius Linn. Sparrow Hawk. The sparrow hawk was recorded from the islands by Colonel Gray- son, but we did not see a single individual, and it probably occurs merely as a straggler. Falco columbarius Linn. Pigeon Hawk. Colonel Grayson records this species as very common upon the islands. Not a single individual was seen by us, and it probably occurs only as a winter visitant or stray migrant. Falco albigularis Daudm. White-throated Falcon. Hypotriorchis rufigularis Grayson, Proc. Boston Soc. Nat. Hist., XIV, p. 269, 1871; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 301, 1874. Colonel Grayson took a specimen of this fine little falcon on the islands. None were seen by us, and it must, no doubt, be classed as one of the numerous accidental visitants from the mainland. Polyborus cheriway pallidus Nelson. Tres Marias Caracara. Polyborus audubonii Grayson, Proc. Boston Soc. Nat. Hist.. XIV, p. 268, 1871, "(part). Polyborus tharus var. auduboni Lawr., Mem. Boston Soc. Nat. Hist., II, p. 303, 1874, (part). Polyboriis cheriway pallidus Nelson, Proc. Biol. Soc. Washington, XII, p. 8, 1898. Very abundant about the settlement on Maria Madre and rather commonly distributed elsewhere over the island. They were also rather common on the other islands, including San Juanito. The old log roads and dry bottoms of the canyons on Maria Madre were favorite resorts. The birds were met in many unexpected places, and were frequently seen perched in tree tops in the midst of the unbroken BIRDS OF THE TRES MARIAS ISLANDS. 39 forest. Iguanas were excessively numerous, and furnished the main supply of food for the caracaras ; the birds were on the alert, however, for anything in the form of carrion that turned up along shore or in the forest. Pandion haliaetus carolinensis (Gmel.) American Osprey. Several ospreys were seen along the shores of all the islands, where they probably nest. Colonel Grayson found a nest on a rock adjacent to the northern shore of Maria Madre and another in a large cactus. Mr. Forrer obtained an immature bird during his visit to this island. Strix pratincola Bonap. American Barn Owl. Colonel Grayson states that he heard the well-known hissing scream of this bird nt night on the Tres Marias. We were told of the pres- ence of barn owls on the islands, but did not see them, and failed to learn whether they occur as residents or stray migrants. Speotyto cunicularia hypogaea (Bonap.) Burrowing Owl. . Colonel Grayson says of the burrowing owl, " A few individuals of this species inhabit the Marias Islands, perhaps wandered from the m -inland." We neither saw nor could learn anything of them, and those seen by Colonel Grayson were probably winter stragglers. ? Micropallas Sp. The first evening after landing on Maria Madre, Mr. Goldman saw and heard a little owl about the size of Glaucidium phalamoides on an old log road in the forest. It was very near, and when he had moved back far enough to shoot without destroying the bird it became too indistinct to see and so escaped. Its notes were not like those of the ferruginous owl. This was the only one seen or heard during our stay. If a resident, it is very uncommon, for special but unsuccessful efforts were made to find others.1 Amazona oratrix Ridgway. Double Yellow-headed Parrot. Chryaotia lemillantii Grayson, Proc. Boston Soc. Nat. Hist., XIV, p. 271, 1871; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 296, 1874. Amazona oratrix Ridgway, Man. N. Am. Birds, p. 594, 1887. The yellow-headed parrot is a common resident and always nests in holes high up in large trees in the forest. Half grown young were found the middle of May. The birds were usually seen flying about the forest in pairs, but congregated in flocks of from six to twenty or more at their feeding places. They feed upon the fruits of various trees, and during May the fleshy pods of Pithecolobium dulce formed their staple article of diet. These low trees, from 15 to 35 feet high, were growing around the houses of the settlement on Maria Madre and were scattered thence along the coast, especially about the borders of 'During the visit to Maria Magdalena Island a larger owl was seen in the forest along the bottom of a steep canyon, but was not sutticiently near to identify, and csc:i]KMl before it could be obtained. This was probably Ciccaba aqitamulata (Bonap.), a species which is not rare on the mainland. 40 NORTH AMERICAN FAUNA. old clearings and in scrubby second growth on the lower slopes. A number of yellow-headed parrots came down every day to feed in the trees, even among the houses, and did not pay the slightest attention to passing people. As these birds readily learn to talk, they are highly prized as pets, and are sold to visitors, or sent to towns on the main- land; the birds taken while young being most highly prized on account of their docility. The men search for their nests, and when one is located the hunter strikes the base of the tree several sharp blows with a stone or ax, and then places his ear against the trunk and listens. He can tell whether the young are old enough to remove, by the strength of the cries they utter in reponse to the blows on the tree. Being satis- fied of the presence of his game, the hunter climbs the tree, and if necessary cuts into the nest with his machete. Each brood contains two young, which are carried to the ground inside the hunter's shirt. By means of a noose on the end of a long cane, like a fishing rod, many old parrots are captured while feeding. An old woman had twenty birds which she had taken in this manner while they were feeding in the top of a small Pithecolobium tree by her door. The hunters search for regular feeding places in the forest and wait under the trees for the birds to come. When the birds arrive, the end of the rod is slowly and cautiously pushed up through the branches, the noose slipped over tlie bird's head and drawn about its neck with a quick jerk, after which the victim is hauled down and thrust into a cage. A favorite resort for the parrots on Maria Madre was a group of trees about half a mile from the settlement. The birds were very unsuspicious, and one could walk up within 20 or 25 yards in full view and watch them without their paying much attention. The parrots were constantly chattering, and the greatest good comradeship seemed to prevail. Mates kept close together and showed their attachment by caressing and feeding one another at short intervals. The proficiency in speaking which some of these birds attain is remarkable. The daughter of the customs inspector on Maria Madre had one which afforded much amusement by the variety of its remarks and their frequently absurd appropriate- ness. Colonel Grayson supposed these birds to be peculiar to tlie Tres Marias, as he did not chance to find them on the mainland. In reality, they are widely distributed on both coasts of Mexico. After comparing the series taken on the islands with specimens from both coasts of the Mexican mainland certain slight differences are noticeable, but are too poorly defined to be worthy of subspecific recognition. The island birds are usually apple green on the dorsal surface, and mainland specimens are more of an oil green; the ventral surface has a more decided bluish wash ; there is also tendency to a richer suffusion of orange and orange red on the yellow feathers about the necks of old birds. The following measurements show that the island birds are a little larger than those of the mainland, with proportionately smaller bill and BIRDS OF THE TEES MARIAS ISLANDS. 41 shorter tarsus. Averages of 9 specimens from the islands (both sexes) : Wing, 233.3; tail, 133.6; culmen, 34.3; tarsus, 24.7. Averages of 7 specimens from both coasts of the mainland (both sexes) : Wing, 222.5; tail, 120.1; culmen, 34.4; tarsus, 24.5. Some old residents on Maria Madre reported that they had occa- sionally seen stray individuals of another parrot which was a little smaller than the yellow head, probably Amazona finschi, which is abundant on the mainland. Psittacula insularis Ridgway. Tres Marias Lovebird. Psittacula cyanopyga Grayson, Proc. Boston Soc. Nat. Hist., XIV, p. 271, 1871; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 297, 1874 (part). Psittacula insularis Ridgway, Proc. U. S. Nat. Museum, X, 1887, p. 541 (Aug. 1888). Psittacula cyanopygia Salvador!, Cat. Birds Brit. Mus., XX, p. 249, 1891 (part). Lovebirds, or 'catarinas,' as they are called by the Mexicans, are common on Maria Madre and Maria Magdalena, and probably occur on Maria Cleofa, although none were seen on the latter island. They were usually seen in flocks, from a few pairs up to 30 or 40 individuals, and in May were feeding on small, sweet, wild figs, common on the lower slopes. While feeding they keep up a constant chattering, which noti- fies one of their presence. When flying over the top of the forest they keep in compact flocks and move steadily forward with rapid wing beats, suggesting a flock of cedar birds. They are very gentle, affec- tionate little creatures and quickly become tame and greatly attached to their owners. Salvadori considers Paittacula insularis a synonym of P. cyanopyga. The series of eight adults from the islands and seven from the main- land show that the island birds can be readily distinguished. The two series show no signs of intergradatiou in color, and their specific distinctness is well defined, although the average measurements show but slight differences in size. Description of Psittacula insularis, $ ad., Maria Madre, May, 1897: Cheeks, lores, forehead, and crown, back to line between orbits, rich green, decidedly clearer than in cyanopyga, and rnnch more sharply contrasted with sur- rounding colors. Middle of crown green, shading abruptly into dull bluish green, which extends thence over neck, middle of back, scapulars, tertials, and lesser wing coverts. Rump patch, axillars, and greater wing coverts vivid cobalt blue. Upper tail coverts brighter green than middle of back, but darker than in cyanopyija. Upper surface of tail dark green, darker than in cyanopyga. Description of Psittacula cyanopyga, $ ad., Tepic, Mexico, April, 1897: Cheeks, lores, and forehead bright green, more yellowish than in insularis, and shading gradually into surrounding colors. Crown, from between orbits, neck, mid- dle of back, scapulars, tertials, and lesser wing coverts, dark green without bluish. Rump patch, axillars, and greater wing coverts bright turquoise blue. Upper tail coverts clearer green than back. Upper surface of tail rich green. 42 NORTH AMERICAN FAUNA. Secondaries ami outer vanes of prima- ries near base, dark blue ; primaries with outer vanes along distal half and at tips edged with dark green; inner webs of primaries brown. Exposed under surface of primaries and secondaries dull blue. Lower parts, back to crissum, dingy blue with a dull greenish wash; blue brightest on abdomen, and contrasting abruptly with green of crissnm. Crissuin rich dark green. Secondaries and outer vanes of inner primaries dark blue ; rest of outer vanes and tips of primaries dark green ; inner webs of primaries brown. Exposed under surface of primaries and secondaries dull bluish green. Lower parts, back to crissum, dull green, richest on abdomen and shading insensibly into color of crissum. Crissum brighter green. Description of Psittacula insularis, 9 ad., Maria Madre, May, 1897. Description of Psittacula cyanopyga, 9 ad., from Tepic, Mexico, April, 1S97. Entire dorsal surface dingy green; Dorsal surface dark green; brightest brightest on sides of head, forehead, on forehead, rump, and upper tail coverts, rump, and upper tail coverts, with a dull with an olive shade on middle of back bluish shade on middle of back and and wings. wings. Lower surface dull green, back to Lower surface nearly uniform light crissum; the latter brighter green. green, with a yellowish shade; crissum nearly the same. Areruye measurements of Psitlacnla insularis and P. cyanopyya. Num- Name. Locality. Sex. ber of speci- Wing. Tail. Cul- ' men. Tarsus. mens. Psittacula insularis.... Maria Madro Island 'ad. J 6 91.8 45.8 13.9 12.9 Paittacnla insularis do ad ? 2 91 5 4C 5 13 5 13 Psittacula cyanopyga. . Tepic and Jalisco ad. tf 3 90. C 45.0 12.8 13.1 do . ad 2 4 89 •> 41 5 13 1 12 7 ' | * Coccyzus minor (Gmel.) Mangrove Cuckoo. This bird is rather common among the mangroves and other trees bordering the salt lagoons near San Bias. A single specimen was taken on May 8 near the shore of Maria Madre, bnt no others were seen, and it probably occurs there only as a straggler. Trogon ambiguus goldmani Nelson. Goldman's Trogon. Trogon ambiguus Grayson Proc. Boston Soc. Nat. Hist., XIV, p. 272, 1871; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 290, 1874 (part). Trogon ambiyuiis goldmani Nelson, Proc. Biol. Soc. Washington, XII, p. 8, 1898. Goldman's trogon was common in the more heavily wooded parts of Maria Madre and Maria Magdalena, and occurs no doubt on the less heavily wooded Cleofa. On .Maria Madre they were found from the coast up nearly to the summit in suitable timber and especially along the sides and bottoms of heavily wooded canyons. Their habits were similar to those of the mainland bird. They sit quietly for a time on a branch BIRDS OF THE TRES MARIAS ISLANDS. 43 arid then fly, with an undulating motion, to another tree in the vicinity. Their notes are limited to a short succession of unmusical sounds, which are frequently heard. They were reported to nest in hollow trees. Unlike most of the birds of these islands, the trogons were nearly as shy as their representatives on the mainland. In life they have light yellow bills and bright red eyelids. Ceryle alcyon (Linn.). Belted Kingfisher. Colonel Grayson records that during each of his visits one or two belted kingfishers were observed sitting on rocks along the seashore. None were seen by us. Dryobates scalaris graysoui (Baird). Grayson's Woodpecker. I'icus scalaris var. graysoni Baird, Hist. N. Am. Birds, II, pp. 515, 517, 1874 ; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 294, 1874. This is the only woodpecker found on the Tres Marias. It is com- mon on all of the islands and generally distributed, except in the most heavily wooded areas. It seems to prefer second-growth thickets and other places where shrubs and scrubby trees form low and rather thin forests, and was usually seen hunting for food along the trunks of large shrubs or small trees. It was a common practice for them to alight on tree trunks near the ground and work slowly to the top, and after remaining there quietly for a time to make a short flight to another tree. Like its relative of the mainland, it is a very quiet bird, rarely uttering any call notes and making little noise while searching for food. From its habit of peering into crevices of the bark and doing only a small amount of pecking it is evident that it finds most of its food on or near the surface. Old nesting sites were seen in the trunks of both living and dead trees and in the flower steins of large magueys. The holes were usually between 5 and 10 feet from the ground. Colonel Grayson found a nest about 12 feet from the ground in the green flower stem of a large maguey (Agave) near the seashore in April. A comparison of a series of these island woodpeckers with other forms shows that the dorsal surface, including the crest of the males, is most like typical D. scalaris from the plains of Puebla. On the ventral surface it may be distinguished from all the other races by its whiter color and scantier and smaller black markings along the sides. These markings are usually in the form of small rounded spots instead of more or less elongated streaks, as in the other races, and the white markings on the greater and lesser wing coverts are decidedly larger and more conspicuous. D. s. graysoni averages a little smaller than D. s. scalaris. The darker dorsal surface and unmarked bases of outer tail feathers distinguish it trom D. s. bairdii. it has a shorter, stouter bill than D. s. lucasanus, with considerably more barring on the outer tail feathers. From D. s. sinaloemis it is distinguished mainly by its larger size, darker dorsal surface, and paler, less marked lower sur- 44 NORTH AMERICAN FAUNA. face. The following averages show the relative size of graysoni and sinalotnsis : Measurements of Dryobates a. graysoni and Dryobates .s. sinaloensis. Num- Name. Locality. Sex. ber of speci- Wing. Tail. Cnl- iin-ii. Tarsus. mens. Maria Madre Island ad.rf 6 98.7 59.6 22.1 18.6 graysoni. do ad. ? 3 96.6 57 19.5 17.5 graysoni. Dryobates scalaris Sinaloa and Tepic sd.rf 3 94.6 53.6 19.8 17.1 sinaloensis. An adult female in the National Museum collection from Mazatlan, while having the normal bill and tarsus of sinaloensis, agrees with birds from the Tres Marias in its long wings and tail. Its measure- ments are as follows: Wing, 96; tail, 60; culmen, 18.5; tarsus, 16.5. Nycticlromus albicollis insularis Nelson. Tres Marias Parauque. Nyctidromus albicollis Grayson, Proc. Boston Soc. Nat. Hist. XIV, p. 273, 1871 ; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 291, 1874. Nyctidromus albicollis insularis Nelson, Proc. Biol. Soc. Washington XII, p. 9, 1898. On the mainland the parauques are rarely seen while the sun is above the horizon, but when night falls they come out of the dense thickets where they have passed the day and sit in dusty trails and other open places. On Maria Madre they were among the commonest birds frequenting old log roads through the forest and shady canyon bottoms until late in the morning and coming out again at 3 or 4 o'clock in the afternoon. Of late years these places have been so completely given over to solitude that when a human being chances to stray into them he is looked upon with little fear. The wood folk seem to con sider him harmless and only a strange creature of their own kind. Parauques were among the most confiding birds found in these quiet retreats and permitted a close approach before taking wing and moving away. In the early dusk they were frequently seen hawking for insects among the low trees. Several came about camp at the north end of Maria Madre just after sunset, and flew very swiftly back and forth with the same erratic course and vigorous wing strokes that are so characteristic of the night-hawk. In fact, I mistook one of these birds for a night hawk until it was secured. Their notes remind one slightly of the whip-poor-will's, but are not so loud and far-reaching. The regular call is made up of two and sometimes three syllables, besides which they have various little clucking and purring notes. Curiously enough the parauques of the Tres Marias bear a much greater resemblance, in size and color, to X. albicollis merrilli of the Eio Grande Valley than to the ordinary birds of the adjacent mainland. BIRDS OF THE TRES MARIAS ISLANDS. 45 Chordeiles acutipemiis texensis (Lawr.) Texas Nighthawk. A single specimen was taken May 5 on Maria Madre, and several others were seen during the first half of the month, after which time they disappeared. These birds were probably stray migrants, for there was nothing to indicate that they were residents. Amazilia graysoni Lawr. Grayson's Humming Bird. Amazilla f/raysoni Lawr., Ann. Lye. Nat. Hist., N. Y., VIII, p. 404, 1867. Pyrrhopha-na yraysoni Grayson, Proc. Boston Soc. Nat. Hist., XIV, p. 283, 1871; Lawr., Mem. Boston Soc. Nat. Hist., II, p. 292, 1874. Very common on the islands. They were seen almost everywhere darting about the less luxuriant parts of the forest hunting for flowers, and now and then stopping on a twig in some low tree top to rest or arrange their plumage. They were among the numerous leathered visitors to the little flower garden at the custom-house on Maria Madre where they were very confiding, and would carry on their search for food among the flowers quite indifferent to one's presence. These hum- ming birds are very pugnacious, as the following incident recorded by Colonel Grayson well illustrates : " Sometimes combats between them become of a desperate nature. One day while watching a number of them in active motion around some tobacco flowers (of which they seem to be very fond) two fine males, after darting at each other for some time, at length came to a deathly struggle, high above my head; they finally clinched each other, each having one of the mandibles of the other in his mouth, at the same time scratching with their little claws, and using their wings with the greatest force, and in this situation, whirling round and round, they fell to the ground near my feet. During this terrible conflict, in which passion and desperation were exhibited, I observed them for a few seconds and then gently placed my hat over both. Even after they were thus captured, and I held one in each hand, they evidenced a desire to continue the war." The same author records having seen these hummers dart upon and capture little flies in the manner of a flycatcher, and found their crops full of minute insects. This I can corroborate from my own observations. Upon comparing a series of 8 specimens of Amazilia graysoni with an equal number of A. cinnamomea the general style of coloration is seen to be very similiar, yet the differences between the two forms are so constant it seems advisable to consider them specifically distinct. The dorsal surface of A. graysoni is a more dingy green with less coppery iridescence than in cinnamomea; the bronze tips of the tail feathers are less uniform, and the extreme points sometimes terminate with a little cinnamon spot; the cinnamon of the lower surface is considerably darker. There is also a well marked and constant difference in size — graysoni being the larger, as shown by the following averages: 46 NORTH AMERICAN FAUNA Measurements of Amazilia yraysoni and A. cinnamomea. Name. Locality. Sex. Num- ber of speci- mens. Winjr- Tail. Cul- men. Amazilia ra soni Maria Madre Island ad. cf 4 68.8 42.6 25 Am 'li ' ni do ad ? 4 65 9 43 4 <>5 4 A T "Western Mexico ad. cf 7 57.9 36.6 22 A ili' cinnamomea do a4G'>2 692 do May 29, 1897 ' Nos. 683-685 were found living in an old house. Ctenosaura terea (Harlan). Black Iguana. The material at hand is very unsatisfactory inasmuch as all the full-grown specimens are of the same sex and in rather poor state of preservation, while the younger specimens afford no characters for 13950— No, 14 5 b'b NORTH AMERICAN FAUNA. satisfactorily separating the various forms which naturally group themselves around Ctcno.saura 1cren. They are therefore left under that general name for the present, the writer hoping some day to be able to review the whole genus. The chief difficulty now lies in the lack of typical specimens of C. teres from Tampico and from the east- ern coast of Mexico generally, and until a series of full-grown speci- mens of both sexes is obtained from that region it will be futile to attempt to straighten out the nomenclature of these lizards. As far as I can make out from my defective material the Tres Marias and Isabel specimens differ sufficiently from specimens from Colima and Telmaii- tepec to warrant their subspecific recognition, but whether identical with the Mazatlau form or not I am not able to say. There are certainly several pretty well defined races of this species; but more adult speci- mens and a direct comparison with the types of many of the old names in various foreign museums will be necessary before the intricate ques- tions involved can be settled. [The females were burrowing in the gravel in dry washes and flats on the islands the last half of May. The burrows were from 2 to 3 or 4 feet deep, and after the eggs had been deposited at the lower end, the female scraped in loose gravel until the hole was filled, and frequently raised a little mound over the entrance. — K. W. X.j U.S. Na- tional Museum number. Collect- ors' num- ber. Locality. Date. 24623 655 Maria Mmlre Island "Slay 4 1S97 °46'?4 656 do Mav 4 18f)7 24625 24626 659 660 Max 5, MOT Mav :,. 1>U7 84827 24628 662 (bis) 662 do do May 5, 1897 Mav 7. 1897 24629 676 do Mav 4 1897 24630 693 Maria dcntii Island Mav '"J 1897 24631 630 Isaliul Island Vpr ->3 LS97 24632 631 24633 632 do Apr. 23,1897 1 Uta latei alls Boulenger. Mr. Xelson remarks that this species lives on stones and driftwood near the border of the woods along the sea beaches. Uta lateralis was based by Bouleuger in 1883 upon specimens from the Tres Marias and from Presidio, near Ma/atlan, collected by Forrer, and specimens from both localities are designated as 'types' in the •* Catalogue of Lizards in the British Museum.' REPTILES OF THE TRES MARIAS ISLANDS. 67 U. S. Na- tional Museum number. Collect- ors' iiuiii- ber. I.malit.v. Date. 24634 «• Maria Mud re Island 24635 642 do . . . May 4 1897 246:iC 643 do 24887 653 .... do. May 13 1887 94638 670 do •>4639 671 do 24640 672 do May °1 1897 24641 673 do M-iy 21 1897 04640 674 do 24643 675 do •>4644 678 do May 25 1807 24645 670 do May 25 1897 Sceloporus boulengeri Stejueger. N. Am. Fauna No. 7, 1893, p. 180, pi. I, figs. 5a-c. This species appears to be smaller than 8. clarkii, of whicb it is the southern representative. A full-grown male (No. 634e) measures only yoiiim from snout to vent. Van Deuburgh's belief that 8. boulengeri "is the same form as Cope's AS. oliyoporus" (Proc. Phila. Acad., 1897, p. 463) is not well founded. The latter is easily distinguished by having only 2 to 3 femoral pores, besides other differences. It is probably identical with 8. horridus. f. S. N;, tioiiiil Museum number. Collect- ors' num- ber. Locality. Sex. Number of pores. Date. 34646 24647 634 634a Isabel Island do c/ad.... 9 Jun 8 9 Apr. 23,1897 Apr 23 1897 94648 634 b do cf adol.. 7 Apr. 23,1897 04649 634 c do ? ad 9 Apr 23 1897 24650 634 d ,lo 9 9 Apr. 23,1897 24651 634 e do cf ad 9 Apr 23 1897 Ciiemidophorus mariarum Giinthrr. Ciifinidophonis mariarum Giinther Biol. Cent. -Am., Rept. p. 28, pi. XX, April, 1885; Boulenger, Cat. L5/ards, Brit. Mus., p. 368, 1885. The swifts from the Tres Marias are essentially alike. Those from Maria Mad re are the largest and possibly also most distinctly marked ; those from the small detached rock off the west side of Maria Cleofa as well as the one from the main island of that name are somewhat smaller. According to Mr. kelson's observation those from the detached islet, which is a bare rock, the nesting place of numerous sea birds, appeared to him paler when alive than those on the other islands, but now, in alcohol, the difference, if any, is very slight. This species, which was originally described by Giiiither from speci- mens collected by Forrer on the Tres Marias seems to be confined to this group of islands. If so, it is the only species of reptile hitherto 68 NORTH AMERICAN FAUNA. collected which is peculiar to these islands. The species has been recorded from the mainland (by Van Denburgh, Proc. Phila. Acad., 1897, p. 463, who identifies "a large number of lizards from Mazatlan, San Bias, and Tepic" with Giiuther's species), but I am satisfied that these records are based upon specimens of C. gularis mexicanus (Peters) which superficially very much resemble the island species. The mis- identification is probably due to the fact that Cope, in his monograph of the genus, overlooked the different keeling of the caudal scales which is the essential character of this species. List of specimens of Cnemidophorus mariarum. U. S. Na- tional Museum number. Collect- ors' num- ber. Locality. Date. 24652 637 Maria Madre Island May 3 1897 24653 638 do May 3 1897 24654 639 do May 4 1897 24655 640 do May 4 1897 24656 24657 644 645 do May 7, 1897 May 7 1897 24658 646 do Miiy 7 1897 24659 647 do May 7 1897 24660 687 Maria Magdalena Island '.... May 28 1897 24661 701 Maria Cleofa Island (outlying rook) May 30 1897 24662 702 do May 30 1897 24663 703 do May 30, 1897 24664 704 do . May 30 1897 24665 705 do May 30,1897 24666 706 Maria Cleofa Island (main island) May 31,1897 Cnemidophorus gularis mexicanus (Peters). The Isabel Island swifts are identical with the mainland form, two specimens of which were brought from San Bias. They are quite dis- tinct from the species on the Tres Marias, which is well characterized by the smaller femorals and the parallel caudals. It is strange that Cope, having had the latter character clearly pointed out by Boulenger, should have referred G. mariarum to C. gularis as a subspecies. List of specimens of Cnemidophorus gularis mexicanus. IT. S. Na- tional Museum number. Collect- ors' num- ber. Locality. Date. 24667 633 Isabellsland April 23, 1897 24668 633a do April 23 1897 24669 633& do April 23,1897 24670 633c do April 93 1897 24671 633d do April 2:! 1897 REPTILES OF THE TRES MARIAS ISLANDS. 69 SEKPENTES. Boa imperator Daudin. This is tlie first record of this species from the Tres Marias. The species is generally distributed through southern Mexico and Central America. Scale rows 73. Litt of specimens of Boa imperator. V. S. Na- tional Museum number. Collect- era' mini- ber. Locality. Date. 24672 648 Maria Madre Island Oxybelis acuminatus (Wietl). A common species occurring all through tropical America from Guayinas, Mexico, south. List of specimens of Oxybelis acuminatus. U. S. Na- tional Museum number. Collect- ors' n urn ber. Locality. Date. 24673 677 Maria Madre Island May 25 1897 Diplotropis diplotropis (Gunther). This species seems to be confined to western Mexico. It was not collected by Mr. Nelson, but there are two specimens in the British Museum collected by Forrer on the Tres Marias (Leptophis diplotropis Boulenger, Cat. Snakes Brit. Mus., II, p. 110). Drymobius boddaerti (Seetzen). A common species distributed over tropical America. In No. 681 the fourth labials on both sides are divided horizontally, so as to suggest a subpreocular. This is an adult male, and is uni- formly colored above, without any markings. The adolescent specimens are uniform above, with a few scales tipped with black; the anterior part of the underside has square blackish spots. The two young ones have above brown, dark-edged, squarish spots, separated by narrow light-colored interspaces. They are marked underneath like the adoles- cent specimens. No. C81, male ad.— Scale rows, 17; ventrals, 183; anal, £ ; caudals, }ff ; supralabials, 9. 70 NORTH AMERICAN FAUNA. List of specimens of Drymobius boddaerti. U. S. Na- tional Museum number. Collect- ors' num- ber. Locality. Date. 24674 652 juv Maria Madre Island Mav 12,1897 24675 654 ad do May 13 1897 24676 658 ad do Mav 14,1897 24677 661 ad do May 15 1897 24679 681 ad Maria Ma^dalona Island May 27 1897 Bascanion lineatum Bocourt. This species is apparently confined to western Mexico. List of specimens of liascanion lineatum. U. S. Na- tional Museum number. Collect- ors' num- ber. Locality. Date. 24680 650 Maria Madre Island May 12 1897 24681 651 do May 12,1897 24682 660 do May 16 1897 Drymarchon corals melanurus (Dum. & Bibr.) Scale rows, 19; ventrals, 205; anal, 1; caudals, £f, aupralabials, 8. Adult male with the characteristic coloring' of this subspecies, which seems to be confined, to Mexico and Central America. List of specimens of Drymarchon corais melanurus. U. S. Na- tional Museum number. Collect- ors' num- ber. Locality. Date. 24683 664 Maria Madre Island "M-iy 18 1807 Lampropeltis micropholis oligozona (Bocourt). Scale rows, 23; ventrals, 230; anal, 1; caudals, f{; temporals, 2 -f 3. Adult male. Thirteen annuli on body, separated by wide, red inter- spaces, without black spots, both on back and belly; all the annuli. complete, including that on neck and throat, which does not touch the parietals; snout white, with black on rostral and anterior nasal. From Boulenger's account it appears that Ferrer's specimens from the Tres Marias are identical. (Cat. Snakes Brit. Mus., II, p. 204.) Distributed over Mexico and Central America. List of specimens of Lampropeltis microphnlis oliyozona. U. S. Na- tional Museum number. Collect- ors' num- ber. Locality. Date. 24684 661 Maria Madre Island May 16, 1K97 REPTILES OP THE TRES MARIAS ISLANDS. 71 Agkistrodoii bilineatus (Giinther). Scale rows, 23 ; ventrals, 138; anal, 1; caudals, 21 + ff-. Adult male. Southern Mexico and Central America to Nicaragua. Li»t of of .Igkiatrodon bilineatux. r. s. N.,. tiiui.-il Museum number. Collect- ors' num- ber. Locality. 24685 707 Maria Mailnt Islam! May 15, 1897 Crotalus sp. ? No rattlesnake was collected on the Tres Marias by Forrer, nor by Nelson, but the latter informs me that he was told of the occurrence of a rattler on Maria Magdalena Island. NOTES ON THE CRUSTACEA OF THE TRES MARIAS ISLANDS. By MARY J. RATH BUN, Second Assistant Curator, Dirision of Marine Invertebrates, U. S. National Museum. Of the four species of Crustacea taken by E. W. Nelson and E. A. Goldman on the Tres Marias Islands in May, 1897, two are identical with forms inhabiting Lower California, one is found in all the warm countries of the world, while the fourth, a fresh-water shrimp, is dis- tributed throughout tropical America. Gecarcinus digueti Bouvier. Gecarcihus digueti Bouvier, Bull. Mus. Hist. Nat., Paris, I, 8, 1895. Maria Cleofa Island. May 30. One large male (Collectors' No. 717). The type and only specimen hitherto collected is from Lower Cali- fornia, and is in the Paris Museum. This species differs from others found on the Pacific coast in its wider carapace, narrower front, longer legs, and in the form of the abdomen of the male. Measurements. Specimen. a 3 1 Exorbital width. Inferior width of front. Length of merus of sec- ond ambulatory leg. Width of same. Length of carpus. Width of same. 4 *S £ Width of same. Length of dactylns. Width of same. Type, d, Lower Cali- 46.3 09 25 9 33 9.7 16 8 17 7.3 24.2 3.7 cf , Maria Cleofa Island.. 70 104 37.5 13 46.3 13.4 22.5 11.5 21.5 10 33.5 5 The measurements of the legs are exclusive of the large spines, and the length given is that of the anterior or superior margin. The penult- imate segment of the abdomen of the male is very wide. Length and distal width, 12 mm.; proximal width, 21.5. Mr. Nelson says of these crabs : On the Tres Marias we found them only on Maria Cleofa, where they were very numerous above high-water mark on the sandy beaches of the low eastern part of the island. They were also living very abundantly in burrows in the soft soil almost everywhere on the slopes of Isabel Island. They are nocturnal in habits, and caused 73 74 . NORTH AMERICAN FAUNA. some annoyance by walking over us at night while we were camped iu their haunts. They began to come out of their burrows as soon as it became twilight in the even- ing. In both localities most of their burrows were found among the scrubby bushes. On Isabel Island they were often seen during the day sitting in the burrows a foot or so from the entrance, but scuttled back to a safe depth when I approached. Ocypode ' occidentalis Stimpson. Ocypoda occidentalis Stiinpsou, Ann. Lye. Nat. Hist. N. Y., VII, 229, 1860. Maria Magdalena Island. May 28. One female (No. 689). Maria Cleofa Island. May 30. One male (No. 699). This much neglected species is distinct, it seems tome, from O. kuhlii de Haan, of which Miers made it a variety. According to the descrip- tion of 0. Jcuhlii given by de Man (Notes Leyden Mus., Ill, 250, 1881), who had the type before him, O. occidentalis differs from it in having a narrower carapace, in the outer orbital angle directed inward and not outward, in the shorter hand, the length of the upper margin of the palm being less than the width, and in having from 18 to 21 tubercles in the stridulating ridge (de Man gives 8 or 10 for kuhlii, while Miers figures 17). The form of the abdomen of the male furnishes excellent characters for the determination of the species of Ocypode. In 0. occi- dentalis the penultimate segment is much wider at its middle than at its proximal end. It is singular that this species is not mentioned in the revisions of the genus by Kingsley, 1880, or by Ortmaun, 1897. Dimensions of a type specimen, U. 8. National Museum. — Male: Length, 40.5 mm.; epibranchial width, 48; exorbital width, 41; length of supe- rior margin of palm, 22.8; entire length .of propodus, 43.5; greatest width, 24. Range. — Type locality, Cape St. Lucas. Also taken at Turtle Uay and San Jose del Cabo, Lower California, by Mr. A. W. Anthony, in. 1896 and 1897. Grapsus grapsus (Linnaeus). "This crab was very abundant on the rocks along the water's edge on the Tres Marias as well as on Isabel Island." (Nelson.) The species is distributed throughout the tropics. Bithynis jamaicensis (Herbst). Maria Magdalena Island. May 27. One adult, 7 young (No. 709). Maria Cleofa Island. May 30. One adult, 3 young (No. 710). "These shrimps were very numerous in a small stream among the hills in the interior of Maria Magdalena, and were also numerous in streams flowing through the hilly parts of the adjacent mainland." (Nelson.) The species is found on the Pacific slope of the continent from Lower California to Ecuador, and on the Atlantic slope from Texas to Rio de Janeiro. The following localities, not before recorded, are repre- sented by specimens in the U. S. National Museum : On the Pacific 1 Ocypode, not Ocypoda, Fabricius, Entom. Sys., Suppl., 312 and 347, 1798; also Entom. Sys., emend, et auct., IV, index, 115, 1796. CRUSTACEA OF THE TRES MARIAS ISLANDS. 75 coast, La Paz, Lower California; Rio Presidio, Siualoa; Rio de Alica, Tepic; Barranca Ibarra, Rio Santiago, Jalisco, and Rio Armeria, Colinia, Mexico; Riode los Platanales and Quebrada Chavarria Golfito (b»th tributary to the Gulf of Dulce), Costa Rica; River David, Chiri- qui, United States of Colombia, 4,000 feet elevation; Guayaquil, Ecuador. On the Atlantic coast, San Antonio, Tex. ; Las Moras Creek, Kinney County, Tex. ; Brownsville, near mouth of Rio Grande, Tex. ; Amixtlan, and Zacatlan, Puebla, Mexico; Escondido River, 50 miles from Bluefields, Nicaragua; Port Castries, St. Lucia, West Indies. The west African form, />. vollenhovenii (Herklots) is no more than a subspecies of B.jamaicensis. It differs only in the slenderer second pair of feet, the carpal and meral joints of which are subequal. The relative lengths of the rostrum and the antenna! scales and peduncles agree with those in some specimens of jamaicensis. The two forms are considered identical by Dr. Ortmann. According to Dr Edward Palmer, B.jamaicensis is much eaten at Colima, and is offered in the market there as a choice article of food, especially on Fridays and Sundays. PLANTS OF THE TRES MARIAS ISLANDS. By J. N. ROSE, Assistant Curator, Division of Plants, U. S. National Museum.1 The Tres Marias, lying about 65 miles off the west coast of Mexico in about 22° north latitude, are among the last of the west coast islands to be studied. All the others have yielded valuable botanical results, but almost nothing has been known of the flora of these islands except in a commercial way. Several botanical expeditions had been planned to explore the islands, but heretofore none had succeeded in reaching them. They are out of the line of traffic, although some of the smaller steamers stop now and then for fuel, and small boats occasionally ply between the islands and San Bias. They are usually visited during the dry season, as it is dangerous to attempt the passage during summer and autumn. Mr. Xelson visited the islands at the very close of the dry season, when the vegetation is at its poorest, and this accounts for the small number of species collected. His collection contains 154 numbers (Nos. 4179 to 4333) and 136 species, mostly from Maria Madre, the largest of the islands, and only a few from Maria Magdalena and Maria Cleof'a. In the subjoined list the plants are from Maria Madre unless otherwise stated. There are no cultivated plants on the islands, except one or two grasses. Fithecolobium dulce, perhaps introduced, is common and much prized for its delicious fruit. The exportation of Spanish cedar (Cedrela sp.) has long been the chief source of income for the islands, but the available supply of this timber is now nearly exhausted. The flora is purely tropical and does not differ essentially from that of the adjacent mainland. Many of the species have not been reported from the mainland opposite, but this is doubtless because the flora is not well known, since these species have been collected either farther north or south. One hundred and twelve species are named below, of which 11 are new. Many of them have a wide distribution in tropical America; all but 6, except the new species, have heretofore been reported from Mexico; 24 range northward into the United States; 64 extend into Central America; 61 into South America; 44 into the West Indies^ and 21 are found in the Old World. Published by permission of the Secretary of the Smithsonian Institution. 77 78 NORTH AMERICAN FAUNA. The following report must be regarded as a preliminary one. The specimens upon which it is based are simply those in fruit or flower at the close of the dry season, a considerable number of which have not been determined specifically and a few not even geuerically. As will be seen from the list below, mostly trees and shrubs were collected, while the herbs, which spring up in great variety during the rainy season, are scarcely represented. The Gamopetahe and Apetahu have been named by Mr. J. M. Green- man, Graminea} by Prof. F. Lamson-Scribner, and Filices by George E. Davenport. The following new species and varieties are based on this collection: jEgiphila pacifica Greeuinan. Kuphorlna sulcarnlea tresmario- Millsp. Beloperone nelsoni Greemnan. (iiUbertia Insularly Rose. UIIXHS pubes'cens Greenman. J'ilocarpus Insularis Rose. Cordia insularis Greenman. Trnixtrti'iiiia maltbya Rose. Erythritta Janata Rose. Zanthosylum innularin Rose. Euphorbia nelsoni Millspangli. Zanthoxylum n than I Rose. ANNOTATED LIST OF SPECIES. Cissatupelos pareira L. Common in Mexico and other tropical countries. May <"> to 25 (Nos. 4233 and 4262). Argemoiie ochroleuca .Sweet. Widely distributed throughout Mexico. Maria Magdalena Island, May 26 to 28, 1897 (No. 4318). Capparis cyiiophallophora L. Found along the coast of Mexico, South America, and the West Indies. May 3 to 25 (No. 4302). Capparis breynia L. Common in Mexico, South America, and the West Indies. May 3 to 25 (No. 4219). Crataeva tapia L. f Perhaps this is the species which has been reported from Acapulco and Mazatlan. May 3 to 25 (No. 4274.) Ternstrcemia maltbya Rose, sp. nov. Tree 3 to 9 meters high; leaves obovate, entire, obtuse, glabrous, thickish, not black-punctate beneath, 5 to 10 cm. long; flowers solitary; peduncles 2.5 to 3.5 cm. long becoming curved, bracteate a short dis- tance below the calyx; sepals 5, orbicular, 8 to 10 mm. in diameter; petals united at base, acute; stamens numerous; fruit (immature) ovate, 20 mm. long, two-celled; seeds red. This species is in all probability Seemau's No. 2148, collected on the road from Ma/atlan to Dnrango and enumerated in the Biologia Centrali-Ainericana without specific name. PLANTS OF THE TRES MARIAS ISLANDS. 79 Collected on Maria Madre Island, May, 1897, by T. S. Maltby (No. 105) and E. W. Nelson (No. 4242) ; by J. N. Eose near Colorao, Sinaloa, July, 1897 (No. 1675). Wissadula hirsutiflora (Pn-sl) Rose. The type of this species came from Acapulco. It is probably com- mon on the west coast, although its distribution and specific limits are not well known. May 3 to 25 (No. 4250). Abutilon reveiitum Watsou. This species extends as far north as Arizona. May 3 to 25 (No. 4203). Hibiscus tiliaceus L. A common tree in most tropical countries. Maria Magdalen a Island, May 26 to 28 (No. 4328a). Melochia tomentosa L. Common throughout tropical America. May 3 to 25 (No. 4205). Guazuma ulmifolia Lam. Common throughout tropical America. Maria Magdaleua Island, May 26 to 28 (No. 4325). Heteropterys floribunda H. B. K. Common in Mexico and Central America. Maria Magdalena Island, May 26 to 28 (No. 4323). Guaiacum coulter! ? Gray. Seemingly common on the west coast of Mexico. Island specimens do not correspond with the form found on the mainland and may represent an undescribed species. May 3 to 25 (No. 4180). Zanthoxyluni insularis Rose, sp. nov. Tree 6 to 20 meters high, thornless; leaves oddly pinnate; leaflets 6 to 7 pairs, opposite, sessile, obovate to spatulate, obtuse or retuse, 2 to 3.5 cm. long, crenate, with large pellucid dots between the teeth and small scattered dots over the surface, glabrous; flowers unknown; fruit small, in a rather compact panicle; pedicels very short; stipe short and thick. Collected by E. W. Nelson on Maria Madre Island. May 3 to 25, 1897 (No. 4278). Zanthoxylum nelsoni Rose, sp. MOV. Tree 7.5 to 20 meters high, thornless( t) ; leaves oddly pinnate; leaf- lets about 6 pairs, distant, opposite, shortly petioled, 5 to 11 cm. long, rounded at base, long-acuminate, creuate, glabrous on both sides, thickly set with pellucid dots; inflorescence in small compact panicles; perianth complete; petals 4 (?); fruit large in dense head-like clusters, not stipitate. A very peculiar species, unlike any Mexican one known to me. Col- lected by E. W. Nelson on Maria Madre Island. May 3 to 25; 1897 (No. 4279). 80 NORTH AMERICAN FAUNA. Pilocarpus insularis Rose, sp. nov. Tree 3 to 6 meters high, glabrous throughout; leaflets usually in threes, some solitary or iu rows, 5 to 7.5 cm. long, retuse at apex, cuneate at base, in the lateral ones more or less oblique; mid vein prominent, lateral veins indistinct below, not very prominent above ; racemes short and compact, 5 to 10 cm. long; fruiting pedicels hori- zontal, 16 mm. long; ovary deeply 4 or 5-lobed or parted, 1 to 4 lobes not maturing. This species is near P. lonyipcs of Mexico, but with somewhat different leaves, more compact inflorescence, etc. Collected by E. W. Nelson on Maria Madre Island. May 3 to 25, 1897 (No. 4307). Amyris sp. May 3 to 25, 1897 (No. 4237). Picramnia sp. A tree 4.5 to 7.5 meters high; flowers said to be greenish, but none with specimens. Much resembling the South American species P. ciliata Mast., but without flowers or fruit. Exact identification is doubtful. May 3 to 25 (No. 4276). Oclma sp. May 3 to 25 (No. 4238). Bursera gummifera Jacq. Common throughout tropical Mexico, Central America, the West Indies, and extending into Florida. May 3 to 25 (No. 4227). Guarea sp. May 3 to 25 (Nos. 4222 and 4230). Trichilia spondioides Swartz. Common in tropical America. May 3 to 25 (Nos. 4214 and 430!)). Ximenia americana L. Common in most tropical countries. May 3 to 25 (No. 4224). Schcepfia schreberi Ginel. Seemingly rare, but has been collected in Mexico and South America. May 3 to 25, 1897 (No. 4271). Hippocratea sp. Maria Magdalena Island, May 26 to 28 (No. 4320). Maria Madre Island, May 3 to 25 (No. 4226). Colubrina arborea Brandegee. Reported from Lower California and the west coast of Mexicr . May 3 to 25 (No. 4213). Cissus sicyoides L. A common species in tropical America. May 3 to 25 (No. 4198). Serjania mexicaiia Willd. A common species in tropical America. May 3 to 25 (No. 4231). PLANTS OF THE TKES MARIAS ISLANDS. 81 PauUiuia sessiliflora Kadi. Heretofore only known from the type specimens collected by Dr. Edward Palmer in the State of Colima, Mexico. May 3 to 25 (No. 4210). Urvillea ulmacea H. 15. K. Common in Mexico and northern South America. May 3 to 25 (No. 4277). Cardiospermum corindum L. A widely distributed species. Maria Magdaleua Island, May 26 to 28 (No. 4328). Crotalaria lupulina ? H. B. K. Perhaps this species, which is common in Mexico, and extends into the United States. May 3 to 25 (No. 4248). Tephrosia sp. May 3 to 25 (No. 4193). Desmodium sp. May 3 to 25 (No. 4287). Erythriiia lanata Rose, sp. nov. A small tree, 4.5 to 7.5 meters high, with a trunk 10 cm. in diameter; brunches glabrous, bearing mostly single infrastipular spines; leaflets triangular, shortly acuminate, 5 to 10 cm. long, 5 to 7.5 cm. broad, glabrous or nearly so. Fio. 1. — Erythrina lanata; a, calyx; 6, banner; c, keel; d, wing; e, stamens; /.ovary. Inflorescence unknown; calyx lauate becoming glabrate, tubular, 10 to 13 mm. long, truncate, one-toothed; banner 68 mm. long, folded, densely white-lanate, rounded at apex; wings (9 mm. long) and keel (10 mm. long) included within the calyx; ovary densely lanate; legume glabrous, 12.5 to 15 cm. long, strongly constricted between the seeds, long-stipitate, attenuate at tip; seeds small (for the genus), nearly orbicular, 6 to 8 mm. long, bright scarlet, with a dark spot at the micropyle. 13950— No. 14 6 82 NORTH AMERICAN FAUNA. The type of this species is Dr. Edward Palmer's No. 129, from Acapulco, Mexico, collected in 1894-95. To this species I would refer specimens collected by Frank Lamb near Villa Union, State of Sinaloa, January, 1893 (No. 428), and flowering specimens by W. C. Wright from the head of Mazatlan River, January, 1889 (No. 1292), and also those collected by J. N. Eose at Kosario. Sinaloa, July 10, 1897 (No. 1592), and July 22 (No. 1822). The latter two specimens are not in flower and their reference here is attended with some doubt. The seeds are larger and the pods less constricted between the seeds. Here also belongs E. W. Nelson's No. 4303 from the Tres Marias, collected May, 1897. I have tentatively referred to this species E. W. Nelson's No. 2099, taken at an altitude of 480 meters, near Santo Domingo, State of Oaxaca, June 18, 1895. It has similar pods, but is described as being but 6 to 12 cm. high and has more bluntish leaflets. Dr. Palmer says this tree flowers in January, and is often used for hedge fences. It differs from all other Mexican species which I have seen in its white lanate banner. Its one-toothed calyx suggests E. rosea, but in the latter the calyx is described as obliquely truncate. Phaseolus sp. Maria Magdaleua Island, May 26 to 28, 1897 (No. 4319). Canavalia gladiata DC. A species of wide distribution, perhaps throughout tropical America. May 3 to 25 (No. 4190). Rhynchosia minima DC. A common Mexican species extending into South America and the United States. May 3 to 25 (No. 4206). Rhynchosia precatoria(?) (H. B. K.) DC. This species has been reported from Acapulco and Panama. May 3 to 25 (No. 4179). Lonchocarpus sp. May 3 to 25 (No. 4310). Ateleia ( ?) sp. Without flowers or named specimens for comparison it is impossible to name this plant definitely. If it belongs to the genus Ateleia it is perhaps A. pterocarpa, the only species known from Mexico. A shrub or small tree 3.5 to 10.5 meters high. May 3 to 25 (No. 41S6). Cassia emarginata L. Common in Mexico, South America, and the West Indies. May 3 to 25 (Nos. 4192 and 4297). Cassia biflora L. Common in tropical America. May 3 to 25 (Nos. 4194 and 4196). Cassia atomaria L. Common in Tropical Mexico and South America. Maria Magdaleua Island, May 26 to 28 (No. 4321). PLANTS OF THE TRES MARIAS ISLANDS. 83 Bauhinia sp. Apparently belonging to the genus Bauhinia, but very unlike any of the Mexican species with which I ain familiar. A vine 6 to 9 meters long; only in fruit. May 3 to 25 (No. 4300). Acacia sp. This appears to be an undescribed species, of which I collected specimens on the mainland. May 3 to 25 (No. 4188). Albizzia occideiitalis Brandegee. Probably the above species, which is found in Lower California and has been reported from western Mexico. May 3 to 25 (No. 4252). Fithecolobium dulce Benth. Common in tropical Mexico and South America. Often cultivated. May 3 to 25 (No. 4285). Pithecolobium ligustrinum Klotzsch. Common in tropical Mexico and northern South America. Maria Magdaleua Island, May 26 to 28 (No. 4314). Conocarpus erectus L. Common throughout tropical America extending into Florida and reported from tropical Africa. May 3 to 25 (No. 4220). Fsidium sp. Tree G to 9 meters high; flowers white; called 'palo prieto.' This species is not represented in the National Herbarium. May 3 to 25 (No. 4306). Casearia corymbosa ( ?) H. B. K. The Tres Marias specimens should probably be referred to this species although our herbarium material seems to represent more than one species. This form is common on the west coast of Mexico and Central America. May 3 to 25 (Nos. 4270 and-4308). Casearia sylvestris Swartz. Widely distributed throughout tropical Mexico, South America, and the-West Indies. May 3 to 25 (No. 4241). Casearia sp. Maria Magdalena Island. May 26 to 28, 1897 (No. 4326). Passiflora sp. May 3 to 25 (No. 4249). Opuntia sp. May 3 to 25 (Nos. 4263 and 4286). Gilibertia insnlaris Rose, sp. nov. Tree 6 to 12 meters high ; leaves 25 to 35 cm. long, including the slender petioles (7 to 18 cm. long), 9 to 20 cm. broad, entire or 3-lobed, oblong, rounded at base, rounded at apex or with a short acumination, glabrous, 3-uerved at base; fruiting inflorescence a short dense panicle; 84 NORTH AMERICAN FAUNA. rays 2 to 3 cm. long; pedicels 4 to 8 mm. long; fruit white, 6-lobcd, 4 mm. high; styles short, connate to near the top. Collected on Maria Madre Island May 3 to 25 (No. 4282). Fortlaiidia pterosperma Watson. A species recently described by Dr. Watson, the type coming from near Guaymas, Souora. May 3 to 25 (No. 4211). Eupatorium sp. May 3 to 25, 1897 (No. 4225). Eupatorium sp. May 3 to 25, 1897 (No. 4244). Eupatorium collinum DC. Common in Mexico and Central America. May 3 to 25, 1897 (No. 4199). Mikania cordifolia Willd. Reported from Central and South America. May 3 to 25, 1897 (No. 4299). Conyza lyrata H. B. K. Reported from Mexico, Central and South America. May 3 to 25 (Nos. 4290 and 4312). Baccharis glutinosa Pers. A common Mexican and Central American plant. May 3 to 25, 1897 (No. 4291). Pluchea odorata Cass. Widely distributed in Mexico and South America. May 3 to 25, 1897 (No. 4181). Parthenium hysterophorus L. Common in Mexico, South America, and in the southern United States. May 3 to 25, 1897 (No. 4267). Perityle microglossa Benth. A common Mexican plant. May 3 to 25, 1897 (No. 4266). Porophyllum iiummularium DC . Eestricted to Mexico. May 3 to 25, 1897 (No. 4292). Trixis frutesceiis P. Brown. A common Mexican and Central American plant. May 3 to 25, 1897 (Nos. 4191), and Maria Cleofa Island, May 30, 1897 (No. 4331). Jacquinia macrocarpa Cav. Species not represented in the National Herbarium, but reported from Mexico, and Central and South America. May 3 to 25, 1897 (No. 4208). Gonolobus sp. Fruit only. May 3 to 25, 1897 (No. 4313a). Buddleia verticillata H. B. K. A common Mexican species. May 3 to 25, 1897 (No. 4183). PLANTS OF THE TRES MARIAS ISLANDS. 85 Cordia soiiorae Rose. A recently described species from Sonora. May 3 to 25, 1897 (No. 4207). Cordia iiisularis Greeninan. Cordia iiixnlaris Crcenman, Proc. Amer. Acad. 33 : 483. 1898. The original description is as follows : "Shrub 3 to 5.5 m. high ; stems and branches glabrous, reddish brown, conspicuously dotted with numerous whitish lenticels; the extreme branchlets covered with hir- sute pubescence; leaves scattered, elliptic-ovate or sometimes slightly obovate, 1.5 to 3 cm. long, 1 to 1.5 cm. broad, narrowed below into a short petiole, obtuse, the upper portion more or less deeply crenate- dentate, occasionally sharply toothed, entire toward the base, hispid above, spreading hirsute-pubescent beneath, especially on the midrib and veins; inflorescence capitulate; heads small (after the corolla has fallen, about 5 mm. in diameter) ; peduncles, during anthesis, 1 cm. or less in length, covered with a spreading hirsute pubescence; calyx 2 mm. long, 5-dentate; teeth short, acute; corolla 3 mm. long, nearly cylindrical, with short recurved lobes, externally glabrous, pubescent inside along the line of the filaments, stamens included; style a little exserted. Collected by E. W. Nelson on Maria Madre Island of the Tres Marias group of islands. May 3 to 25, 1897 (No. 4296)." Tournefortia Candida Walp. Not previously in herbarium. May 3 to 25, 1897 (Nos. 4217 and 4229). Tournefortia cymosa L. I have only seen specimens from Guatemala. May 3 to 25, 1897 (No. 4189). Tournefortia velutina H. B. K. Reported from the west coast of Mexico and Guatemala. May 3 to 25, 1897 (No. 4209). Heliotropium indicum L. Common in Mexico and most tropical countries. May 3 to 25, 1897 (No. 4253). Heliotropium curassavicum L. Common in Mexico and South America as well as in the Old World. Reported in the United States as far north as Oregon and Virginia. May 3 to 25, 1897 (No. 4313). Ipomoea bona-nox L. A common tropical plant extending into Florida. May 3 to 25, 1897 (No. 4269). Ipomoea peduncularis JJertol. Common in Mexico and Central America. May ,'t to 25, 1897 (No. 4235). 86 NORTH AMERICAN FAUNA. Jac quern on tia violacea Choisy. Beported from Mexico, Central and South America, and the West Indies. May 3 to 25, 1897 (No. 4251). Solanum nigrum L. A widely distributed species. May 3 to 25, 1897 (No. 4200). Solanum lanceaefolium Jacq. A common tropical plant. May 3 to 25, 1897 (No. 4240). Solanum callicarpaefolium Kunth & Bouch6. Common in south Mexico and northern South America. Maria Magdalena Island, May 26 to 28, 1897 (No. 4322). Solanum torvum Swartz. Common in Mexico and Central America. May 3 to 25, 1897 (No. 4185). Solanum verbascifolium L. Only reported hitherto from one station in southern Mexico. May 3 to 25, 1897 (No. 4216). Fhysalis pubescens L. A common tropical plant. May 3 to 25, 1897 (No. 4255). Bassovia donnell-smithii Coulter. A recently described South American and Guatemalan species. May 3 to 25, 1897 (No. 4232). Datura discolor Bernh. Reported from Mexico and West Indies. May 3 to 25, 1897 (No. 4197). Nicotiana trigonophylla Dtfn. Common in Mexico. May 3 to 25, 1897 (No. 4212). Russelia sarmentosa Jacq. A common Mexican and Central American species. (May 3 to 25, 1897 (No. 4289). Capraria biflora L. A widely distributed plant, extending into Florida. May 3 to 25, 1897 (No. 4195). Bignonia aequinoctialis L. (B. sarmentosa Bertol.) Recently collected at Acapulco by Dr. Edward Palmer. It is com- mon in Central and South America. May 3 to 25, 1897 (No. 4301), and Maria Magdalena Island, May 26 to 28 (No. 4324). Beloperone nelsoni Greenman. Beloperonc nelsoni Greenman, Proc. Amer. Acad. 33 : 488. 1898. " It is nearest B. comosa Nees, in DC. Prodr. 11 : 416, but differs very markedly in the size of the flower and the character of the lower lip." — Greenman in litt. May 3 to 25, 1897 (No. 4246). The original description is as follows: '-'Erect; stems branching, sub- PLANTS OF THE TRES MARIAS ISLANDS. 87 terete, covered with a spreading or slightly retiexed grayish pu- bescence; leaves ovate-lanceolate or oblong-lanceolate, 5 to 10 cm. long, 2 to 4 cm. broad, obtuse at the apex, entire, narrowed below into a slender petiole, densely lineolate above, pubescent on either surface, especially on the veins, later becoming glabrous; petioles about 2 cm. in length; inflorescence terminating the stem and branches in rather close bracteate spikes; bracts oblong or obovate; bracteoles linear, nearly 1 ein. long, exceeding the calyx; calyx about 5 mm. long, deeply 5-parted; divisions nearly equal, lanceolate, acute, ciliate; corolla 2 to 2.5cm. long; tube exceeding the limb; upper lip shortly 2-lobed, the lower more deeply 3-lobed, rather broad, somewhat plaited in the throat; capsule 10 to 12 mm. long, pubescent. Collected by E. W. Nel- son on Maria Madre Island of the Tres Marias group of islands, 3-25 May, 1897 (No. 4246). "A species closely resembling B. comosa Nees, but with a much shorter corolla, and broader lower lip. The leaves are also somewhat larger, longer-petioled, and much less pubescent. It may be that fur- ther material will prove this to be a variety of B. comosa Nees, but as the material at hand shows no sign of intergradation, it seems best for the present at least to regard Mr. Nelson's plant as a distinct species." Laiitana horrida H. B. K. lleported from both northern and southern Mexico. May 3 to 25, 1897 (No. 4187). Citharexylum affinis D.Don. This is a rare Mexican species which has been " compared with the Prodromus specimen at Geneva by 0. De Candolle" — J. M. G. May 3 to 25, 1897 (No. 4311). Greeninan. ifica Greeuman, Proc. Amer. Acad. 33:485. 1898. The original description is as follows : " Shrub 2.5 to 7 m. high ; stems and branches terete, covered with a grayish brown bark and dotted here and there with lenticels, glabrous ; branchlets terete, somewhat compressed at the nodes, fulvous-pubescent; leaves opposite, oblong- ovate, 5 to 15 cm. long, 3.5 to 7.5 cm. broad, more or less acuminate, entire, rounded or rather abruptly narrowed at the slightly unequal base, glabrous, or at least glabrate above, with scattered, tawny, sub appressed hairs beneath, especially upon the midrib and veins; petioles less than 1 cm. in length; inflorescence terminating the stems and branches in rather close paniculate cymes; peduncles, pedicels, the subulate bracts and calyx covered by a fulvous subappressed pubes- cence; calyx about 4 mm. long, 4-lobed; lobes broader than long, submucronate, greenish; corolla tubular, 10 to 12 mm. long, glabrous; tube somewhat ampliated above; lobes oblong-elliptic, about 4 mm. long,obtuse; stameusequal or rarely unequal, exserted; filaments pubes- cent below, glabrous above; drupe yellow, obovoid, 8 to 10 mm. long, 0 88 NORTH AMERICAN FAUNA. to 8 mm. in diameter, one-half or more exserted from the persistent coriaceous subcrenately lobed, cup-shaped calyx.— Collected by E. W. Nelson on Maria Madre Island of the Tres Marias group of islands, 3-25 May, 1897, No. 4245 (in flower) and No. 4254 (in fruit)." Hyptis albida H. B. K. Several times reported from Mexico. May 3 to 25, 1897 (No. 4223). Salvia aliena Greene. A Mexican species. May 3 to 25, 1897 (No. 4247). Stachys coccinea Jacq. Common in Mexico and Central America, extending into Texas and Arizona. May 3 to 25, 1897 (No. 4265). Iresine interrupta Benth. Reported from western and central Mexico. May 3 to 25, 1897 (No. 4234). Phytolacca octandra L. May 3 to 25, 1897 (No. 4293). Stegnosperma halimifolia Benth. Common along the west coast of Mexico. May 3 to 25, 1897 (No. 4184). Batis maritima L. Extending from Florida and California to Brazil and the West Indies and also reported from the Sandwich Islands. Magdalena Island, May 26 to 28, 1897 (No. 4327). Coccoloba leptostachya Benth. This species has not been heretofore found in Mexico, but has been reported from Central America and South America. Maria Magdalena Island, May 26 to 28, 1897 (No. 4315). Antigonon leptopus Hook. & Am. A very common vine on the west coast of Mexico. May 3 to 25, 1897 (No. 4204). Aristolochia par din a Dnch. A little-known plant collected at Colima many years ago by Ghies- brecht, and recently at the same place by Dr. Edward Palmer. May 3 to 25, 1897 (No. 4304). Piper aduncum L. Reported from Mexico, Central and South America, and the West Indies. May 3 to 25, 1897 (No. 4283). Euphorbia sp. May 3 to 25, 1897 (No. 4268). Euphorbia subcaerulea tresmariae Millspangli, var.nov. "In the characters present in the specimens collected, this agrees well with E. subcaerulea .Rob. and Greenm.(Pringle No. (5265, Oaxaca), except in the hairy involucre more regularly toothed involucral lobes, and in PLANTS OF THE TRES MARIAS ISLANDS. 89 having the styles bifurcate, to the middle only, and flat spreading with no tendency to reflexion or peltation as in the other species. The fruits may prove this to be a distinct species. May 3 to 25, 1897 (Nos. 4298 and 4202)."— Millspaugh MSS. Euphorbia sp. Specimens are indeterminable from lack of characters. May 3 to 25, 1897 (No. 4215). Euphorbia nelsoni Millspaugh. Euphorbia nelsoni Millspangh, Bot. Gaz.26:268. 1898. May 3 to 25, 1897 (No. 4294, not 4284, as published). FIG. 2. — Euphorbia neltoni. The original description is as follows: "Fruticosa, glabra, longe et corymbosa rarnosa, ramis teretis, internodiis longis, cortex maculatis, maculio oblongis roseus. Foliis inferioris fasciculatis, petioliis longis filamentosis, pagina tennis ovato-cuneatis, obtusis, apiculatis, foliis floralibus oppositis, orbiculatis petiolis limbum aequantis. Involucriis terminalibus corymbosis, pedunculatis, campanulatis glabris, lobis latis truncatis irregulariter C-8 fimbriatis, glandulis 5, transversis oblongis integris, appeudicibus minntis vel nullus. Stylis longis revoluto-cir- cinalis. Capsule luridre profunde tri-sulcato, semine sub-globosis pallide-fuscis, scrobiculatis, linea media nigra geminatis, ruga? anasto mosautis tuberculatis 2 mm. long, 1.9 mm. lat." Several Euphorbias were collected on the islands in too imperfect condition to determine, and it has been thought advisable to reproduce the cut1 of the present species for the purpose of assisting future study of the flora. 'Through the kindness of the editors of tho Botanical Gazette I am permitted to .use this illustration. 90 NORTH AMERICAN FAUNA. The main figure shows a cluster of leaves. To the right is ti flower cluster and to the left a dissected flower with end and side views of the seed. Garcia nutans Rohr. Found in Mexico and South America. May 3 to 25, 1807 (No. 4228). Croton ciliato-glandulosus Ort. May 3 to 25, 1897 (No. 4218). Acalypha sp. May 3 to 25, 1897 (No. 42GO). Celtis monoica Hemsley. May 3 to 25, 1897 (No. 4230). Buxus pubescens Greenman. Buxus pubescens Greenman, Proc. Anier. Acad. 33 : 481. 1898. The original description is as follows: " Shrub or small tree, 4.5 to 8 m. high; stems and branches covered with a grayish bark; the branchlets and younger shoots provided with a soft, spreading pubes- cence; leaves opposite or occasionally subaltemate, sessile or nearly so, rhombic-ovate to oblong-ovate, 2 to 5 cm. long, 1.5 to nearly 3 cm. broad, 3-nerved, obtuse or acutish, mucrouate, cuneate at the tyase, entire, ciliate, soft-pubescent beneath, more sparingly pubescent and glabrate above, showing the reticulate venation on the upper surface; inflorescence of axillary short-pedunculate much contracted subrace- mose pubescent clusters ; staminate flowers pedicellate; pedicels 3 mm. long, about twice exceeding the ovate acute bracts; calyx deeply 4- parted; divisions ovate, acute, 2 mm. long, the inner divisions slightly broader than the outer ones; the rudimentary pistil somewhat quatre- foil or X-shaped; fertile flowers about 5 mm. long, single, sessile, terminating the inflorescence; ovary glabrous; fruit not seen. — Col- lected on Maria Madre Island by E. W. Nelson, 3-25 May, 1897, No. 4221. "A species apparently endemic in the Tres Marias Islands, and most nearly related to the West Indian B. pulchella Baill." Ficus radulina Watson. A recent species of Dr. Watson's from northern Mexico. May 3 to 25, 1897 (No. 4261). Ficus fasciculata Watson. Only known from western Mexico. May 3 to 25, 1897 (No. 4288). Ficus sp. May 3 to 25, 1897 (No. 4182). Myriocarpa longipes Liebm. Found in Mexico and Central America. May 3 to 25, 1897 (No. 4275). Agave sp. Six meters high, leaves 9 to 18 dm. long; marginal teeth small, dis- tant; end spine short, stout, pungent; capsules oblong, large, 7 cm. long. PLANTS OP THE TRES MARIAS ISLANDS. 91 This species belongs to the subgenus Eugave and the Rigidae group of Mr. Baker's revision. It is near A. vivipara, and perhaps not dis- tinct. Mr. Nelson's plant does not seem to differ from specimens col- lected by me on the mainland. May 3 to 25, 1897 (No. 4264). Cyperus ligularis L. Keported from Mexico, Central and South America, and West Indies, as well as Africa and Australia. Maria Oleofa Island, May 30, 1897 (No. 4330). Cyperus incompletus Link. Eeported from Mexico and Brazil. May 3 to 25, 1897 (No. 4259). Panicum brevifolium L. May 3 to 25 (No. 4257). Eleusiiie indica Gaertu. May 3 to 25, 1897 (No. 4305). Dactyloctenium aegyptiacum Willd. May 3 to 25, 1897 (Nos. 4295 and 4256); Maria Magdalena Island, May 26 to 28 (No. 4317). Arundo donax L. Maria Cleofa Island, May 30 (No. 4332). Zamia loddigesii ( f ) Miq. Reported from Mexico. Maria Cleofa Island, May 30, 1897 (No. 4329). Pteiis longifolia L. Maria Madre Island, May 3 to 25, 1897 (No. 4201). Aspidium trifoliatum Swartz. Maria Madre Island, May 3 to 25, 1897 (No. 4280). Aspidium patens Swartz. A widely distributed species. Maria Magdalena Island, May 26 and 28, 1897 (No. 4316). Adian turn concinnum H. B. K. Maria Madre Island, May 3 to 25, 1397 (No. 4273). Adiantum tenerum Swartz. Maria Madre Island, May 3 to 25, 1897 (No. 4281). Gymnogramme calomelanos Kaulr. A widely distributed species. Maria Cleofa Island, May 30, 1897 (No. 4333). PARTIAL BIBLIOGRAPHY OF THE TRES MARIAS ISLANDS. By E. W. NELSON. 1703. DAMPIER, WILLIAM. A New Voyage round the World, 5th ed., I, pp. 26o-264. Notes the presence of seals about the Tres Marias. These notes are quoted by Allen and Alston. 1865. BAIRD, SPENCER F. , Formicarido', Tyrannidce, and Trochilidw. 05; I'espertiUo n'ujri- cans, p. 206; Mawotus icaterhousii, p. 207; Chccronycteris mexicana [=(;7osso- phaga mutica], p. 207; Procyon cancrii-tfrus, p. 208; Lepus graysoni, p. 211. 1885. MADARASZ, JULIUS VON. Oruithologiai Kozlemdnyek A Magyar Nemzeti Muzeum Gyujtemeny6bol. < TeruK-szetrajzi Fiizetek, IX, p. 74, Feb. 20, 1885. Description of Vireo forreri. 1887. RIDGWAY, ROBERT. < Manual of North American Birds, 1887. Descriptions of lacne laicrencei (ex Berlepsch MS.), p. 320; Platypsaris insularis, p. 325; Piranga flammea, p. 457. 1887. RIDGWAY, ROBERT. A Review of the Genus Psittacula of Brisson. <[ Proc. U. S. Nat. Museum, X, 541. 1887. Description of Psittacula insularis. 1891. BRYANT, WALTER E. Andrew Jackson Grayson. < Zoe. II, pp. 34-68, 1891. A short account of Grayson's life, with extracts from his journals on the habits of certain birds of the Tres Marias. 1898. EVERMANN, BARTON WARREN. Notes on Fishes Collected by E. W. Nelson on the Tres Marias Islands and in Sinaloa and Jalisco, Mexico. <^ Proc. Biol. Soc., Washington, XII, pp. 1-3, 1898. Records Agonostomus nasutus Giinth. on Maria Magdalena and Maria Cleofa. 1898. MERRIAM, C. HART. Mammals of Tres Marias Islands off Western Mexico. . 1898. GREENMAN, JESSE M. Diagnoses of New auardiiia, 88. Arunis, 63, 69. Dove, Mexican Ground, 37. Mourning, 36. Tres Marias, 30. White-fronted, 36. White-winged, 36-37. 1 >r\ marchoii i-orais nielauurus, 70 INDEX. Drymobius boddwrti, 69-70. Dryobatea acalaris, 43. acalaris baircli, 43. scalaris graysoni, 12, 2'i, 43-44. scalaris lucasauus, 43. scalaria sinaloensia, 43-44. Egret, American, 33. Elaiuea plat-ens, 50. Eleusineindica,91. Empidonax difficilis, 49. Erythrina lanata, 13, 78, 81 Eupatorium, 84. collmum, 84. Euphorbia, 12, 88, 89. nelsoni, 13, 78, 89-90. subcoerulea tre.smariae, 13, 78, 88-89. Faleoalbigularis,38. . columbarius, 38. peregrinus anatum, 3S. peregrinus nigrieeps, 38. sparverius. 38. Falcon, White-throated, 38. Ficus t'asciculata, 90. radulina, 90. Florisuga inellivora, 62. Flycatcher, Arizona Crested, 48. Beardless, 49. Golden Crowned, 50. Little Golden Crowned, 50. Olivaceous. 48-49. Western, 49. Fregata aquila, 33. Garcia nutana, 90. Gecarcinus digueti, 73-74. Gilibertia iusularis, 13, 78, 83-84. Glandina turris, 11. Glossophaga unities, 11, 18-19. Glossophaga, Tre.s Marias, 18-19. Goldfinch, Mexican, 52. Gonolobus, 84. Grackle, Great-tailed, 52. Grauatellus fraucescae. 12, 22, 56-57. veuustus, 57. Grapsus grapsus, 74. Guaiacum coulter!, 79. Guarea, 80. Guazuuia ulmifolia, 79. Gull, American herring, 23. Heennann's, 23-24. Gymnogramme calomt-lanos, 91. Hadrostomus aglaiie alh'uis, 47. Htematopus frazari 34-35. galapagensis, 34-35. palbatus. 34-35. Haliplana fuliginosa crissaiis, 24. Hawk. Duck, 38. Pigeon, 38. Sparrow, 38. Tres Marias Bed-tailed, 37, 38. Heliotrppiiim curassavicum, 85. indicum, 85. Heron, Great Blue, 33. Snowy, 33.. Yellow-crowned Night, 33, 34. Heteropterys floribunda, 79. Hibiscus tiliaceus, 79. Hippocratca. 80. Hirundo ervthrogaster, 54. Humming Bird, Gray son's, 45, 46. Lawrence's, 46, 47. Hylocichla ustulata, 60. ustulata almre, 60. ustulata swainsoni, 60. Hypotriorchis rutigularis, 38. llyi.tis albida, 88. laclie latirowtris, 46, 47. lawrencei. 12, 22, 46. Icterus graysoni, 12, 22, 50-51. pustulatus, 51. Iguana, Black, 65, 66. lponio;a, 12. bona-nox, 85. peduucularia, 85. Ircsine interrupta, 88. Jacquemontia violacea, 86. Jacqiiinia macrocarpa, S4. Jay. BeecheyX 50. Kingbird. Couch's, 48. Kiiigtisher, Belted, 43. Kiuosternou iiitegrum, 64. Lamellaxis, 11. Lampropeltis mioropholu oligozoua, 70. Lantana horrida, 87. Larus argentatus .sinithsoniauua, 23. heermanni, 23-24. Lasiurus borealis mexicaiia, 15, 19. Leptophis diplotropia, 69. Leptotilaalbifrons,30. capitalis, 12. 22, 36. fulviventris brachyptera, 36. Lepus graysoui, 11, 16-17. Lonchocarpus, 82. Lovebird, Tres Marias, 41-42. Man o'- war bird, 33. Marmosa insularis, 11, 15-16. Melanerpes, 23. Melanotis caerulescens, 59. cierulescens longirostris, 12, 22, 59. hypoieoooB, 59. Melochia tomentoaa, 79. Mclopelia leucoptera, 22, 36-37. Merula tlavirostris, 60, 61. grayi, 62. graysoni, 12, 22, 60-62. tristis, 62. Micropallas, 39. Mikania«;ordifolia,84. Mimus polvglottos, 57. Mockingbird, 57. Momotus, 23. Mouse, Tres Marias, 16. Murrelet, Short-billed, 23. Mus rattus, 16. Myadestes obscurus, 59. obsfiirus iuaularis, 12, 22, 59-60. obscurua occidentalis. CO. Myiarchus lawrencei olivascens, 48-4'J. mexicanus inagister, 48. Myiopagis placens, 50. placeus niinimus, 12, 22, 50. Myiozetetea,23. Myotis nigricans, 15, 18. Myriocari>a longipes, 90. Ni'cotiaua trigonophylla, 86. Nighthawk, Texas, 45. Nycticorax violacens, 33-34. Nyctidronius albtoolliB,«. albicollis insularis, 12, 22, 44. albicollis merrilli. 44. Oceanodroma mclania, 28. Ochna, 80. Ocv]>odo,74. Or\|M,(lckuhlii,74. occidentalis, 74. Opeas subula, 11. Opossum, Tres Marias Pigmy, 11, 15-16. Opuntia,12,83. Oriole. Grayson's, 50. Ornithion imberbe, 22, 49. imberbe ridgwayi,49. Orthalicus undatus, 11. undatus melanocheilus, 11. Oryzoinya nelsoni, 11, 16. Oaprey, 39. Otopterus mexicanus, 18. Owl, Barn. 39. Burrowing. 39. Oxyliclis acuminatua, 69. Oyster-catcher, American, 34-35. I'andion halia-tus rarolmensis, 39. Panicum brevifolium,91. Parauque, Tres Marias, 44. Parrot, Double yellow-head, 39-41. Parthenium hysterophorus, 84. Parula, iusularis, 55. Tres Marias, 55. Passittora. 12, 83. Paullinia seaailittora, 81. Pclecanus californicus, 32-33. Pelican, California Brown, 32-33. I'etasophora thalassina, 62. Perityle microglossa, 84. Peroiiiyscus madreusis, 11, 16. Petrel.' Black. 28. Phfethon a4h.-reus, 28-29. Phalacrocorax, 32. l'ha.scoliis,82. Phofiuua cuiumuuis, 19. INDEX. 97 Phyllodactylus tubereuU.siis, 63. 64-65. Physalis pubeseens. 86. Pbytolaooa octaiidra, 88. Piaya. 23. Picramnia, 80. ciliata. 80 Pious scalaris graysoni, 43. I'i-i-on. Tres Marias. 35. Pilocarpus insularis. 13, 78. 80. Piper ad ii Hi-mil. 88. Piptto,23. Pirauga bideutata, 53, 54. bidentata flammea, 12. 22, 52. 53, 54 ludovieiana, 23. 52. Pitheeolobhtm, 12. dulce.39,77,83. ligustrinum, 83. Platypsaris aglaia-, 47-48. aglai;e, albiventris, 47,48. aglaias insuiaris. 12. 22, 47-48. aglai.t smnicbrasti, 47-48. Plover, Semipalmated. 34. Plucbea odorata, 84. Polyborus audubonii, 38. aiidubonii insularis, 22. cheriway ].allidus. 12,22,38-3'J Polygyra ventrosula, 11. PoropByUoni nommnlariom, 84. Porpoise, Common, 10. Long-nosed, 19. Portlaudia pterosperma, 84. Procyon lotor iusularis, 11, 17. Prodelphinus longirostris. 19. Psidium. 35, 83. Psittacnla cyanopyga, 41-42. insularis, 12. 22* 41-42. Pterislongifolia,91. Putfinus cuueatus, 27-28. knudseni, 27. Pyrgisoma, 23. Pyrrhopha-na graysoni, 45. Haeooon, Tres Marias, 17. Kat, Black, 16. Xejson's Rice, 16. RhogeSaM parvnla. 11, 18. Rbogeessa, Tres Marias, 18. Rhynchosia minima,. 82. precatoria, 82. Russelia sarmeutosa, 86. Qtiiscalus niacrourus, 52. Saltator. 23. Sal via aliena, 88. Sandpiper, Spotted. 34. Si-eloporns boulengeri. 67. clarkii,67. borridus, 67. olijjoporus, 67. Scbctptia schreberi, 80. Sea Lion, 15, 17-18. Si-rjania mexicana, 80. Shearwater, \Vedjie-tailed, 27-28. Solanum, 12. callicarpaefolium, 86. lanceitfolium, 86. nigrum, 86. torvum, 86. verbascifolium, 86. Solitaire, Tres Marias. 59-60. Spanish cedar, 7, 8, 9, 12. Speotyto cunicularia hypoga-a,39. Stac-hys coccinea. 88. Stegnosperma lialituitblia, 88. Sterna i-lcgans, 24. fiiliginosa crissalis, 24-26. galericulata, 24. maxima, 24. Strix pratincola, 39. Sula ba.ssanii, 29. brewsteri, 29-30. ijiissi. 31, 32. Sula uebonxii, 31-32. ]>is«-ator,31. websteri. 29. Swallow, Barn. 54. Tanager. Louisiana, 52. Tres Marias, 52, 53. Tej.brosia, 81. Tern, Elegant. 24. Pacific Noddy, 26-27. Pacific Sooty, 24-26. Royal. 24. Terflostrcemia maltbya, 13,78. Thalurauia glancopis, 62. lueiie, 62. Tbrush, Olive backed, 60. Russet-backed, 60. Thryothorus felix, 22. 57, 58. lawreiicii, 12, 22. 57, 58. lawreucii magdalenae, 12. 22. 58. Tournefortia Candida, 85. cymosa, 85. velutiua, 85. Totanus flavipes, 34. Tres Marias, Blue Mockingbird, 59. Caracara, 38-39. Cardinal. 52. Chat-Warbler. 56-57. Cottontail. 16-17. Dove, 36. Glossopbaga. 18-19. Lovebird. 41-42. Mouse, 16. Parauque, 44. Parula, 55-56. Pigeon. 35. Pigmy Opossum. 11, 15-16. Raccoon, 17. Red-tailed Hawk, 37-38. Rhogeessa, 18. Robin, 60-61. Solitaire, 59-60. Tanager. 52. Vireo, 54-55. Trichilia spondioides, 80. Trixis frutescens. 84. Trogon ambigiius, 42. ambiguus goldmani, 12, 22. 42-43. Trogon, Goldman's, 42-43. Tropic Bird, Red-billed, 28-29. Turdus flavirostris, 60. Turtle, Mud, 11. Tyrannus melaucholicus couchi. 22, 48. L1 rauomitra guatemalensis, 62. Urvillea ulmacea, 81. Uta lateralis, 66-67. Vireo flavovindis forreri, 12, 21, 22, 54, 55. hypochryseus, 54. hj-pochryseus sordidus, 12, 22, 54-55. A'ireo, Ferrer's, 54. Tres Marias. 54-55. Vulture, Turkey. 37. Warbler, Alaskan Yellow, 56. Audubon's, 56. Pileolated, 57. Townsend's, 56. Tres Marias Chat-, 56, 57. Western Yellow, 56. Wilsonia pusilla pileolata, 57. Wissadnln hirautifiora, 79. Wonil IVwee. Western, 49. \V]-i-u.Mamlali-na,58. Maria Madre, 57-58. Xinienia americana, 80. Yellow-legs, 34. Zalophns californiauus, 17-18. Zamia loddigesii, 91. Zanthoxylum insularis, 13, 78. 70. nelsoni, 13,78,79. Zenaidura macroura, 36. X0. 14- Date Due CAT. NO. 24 161 UTHERN REGIONAL LIBRARY FACILITY AA 000 671 304 4