Thamnophis eques obscurus, CONANT, 2003

CONANT, ROGER, 2003, Observations on Garter Snakes of the Thamnophis eques Complex in the Lakes of Mexico’s Transvolcanic Belt, with Descriptions of New Taxa, American Museum Novitates 3406, pp. 1-64 : 29-37

Thamnophis eques obscurus ,

new subspecies Figures 10 , 11

HOLOTYPE: AMNH 87543 , a moderately large male from the town of Chapala , Jalisco, Mexico, collected August 29, 1961, by Rog­ er and Isabelle Hunt Conant. Total length 762 mm, tail length 179 mm, tail/total length = 0.235.

PARATYPES: All from or near settlements bordering the Lago de Chapala. Chapala AMNH 82032–82046 , 87544 , 87545 , 91693–91695 ; Jamay AMNH 19544 , 19545 ; Jocotepec AMNH 87546 , 87547 , 96841– 96845 ; 10 km northeast of Sahuayo AMNH 93958 .

ETYMOLOGY: From the Latin obscurus in reference to the faintness to complete absence of pale longitudinal stripes.

DIAGNOSIS: A virtually stripeless race of Thamnophis eques . Adults in life lacked pale longitudinal dorsal stripes (fig. 11). A few preserved for many years show faint traces of lateral stripes that now appear bluish. There is no indication of a middorsal stripe in any adult of the sample available. In the young, arbitrarily considered as all specimens less than 300 mm in total length, there may be traces in life of pale stripes immediately posterior to the head.

The venter is grayish and the anterior edge of each ventral scute is black or dark gray (fig. 11). The resultant crossbanding is prominent, but the components are narrower than those on the venters of snakes from the Cuenca de Michoacán. The chin, throat, and underside of the tail are white and more or less unmarked. The anal plate is undivided and usually white in color, but it is dark in a few specimens.

DESCRIPTION OF THE HOLOTYPE: Scutellation the same as that of The Species Model, with the following exceptions: Postoculars 4 on left side of the head instead of 3. Three temporals in second row (on left); 2 on right, but with a small scale about as large as the lower secondary temporal wedged between the anterior temporal and the parietal. Scale rows 21­19­17, reducing to 19 by the loss of the fifth row at the level of ventral 81 on the left and the loss of the fifth row at the level of ventral 77 on the right. Reducing again to 17 by the loss of the fourth row at the level of ventral 94 on the left and ventral 98 on the right. Total ventrals 158; subcaudals 73. Both hemipenes everted and of similar shape and appearance as in figure B in Rossman et al. (1996: 32). A summary of scale counts and other data for the entire sample from the Lago de Chapala is in table 3 (also see the Discussion). One of the 31 specimens, AMNH 93958 from near Sahuayo, was found DOR and was so badly damaged that few data could be recorded from it. Some of the others had incomplete tails.

The holotype was preserved in the field,

TABLE 3 Variation in Scutellation and Tail Length Proportions in Thamnophis eques obscurus , New Subspecies

unfortunately just before unsuspected ecdysis. I had selected a large male as the type, but when I reexamined it in 2000 two soft spots had developed, probably because formaldehyde had not been injected evenly by the person who preserved it. Although the holotype had shed its scales (the stratum corneum) in part, many of the dorsal scales remaining in situ show evidences of apical pits.

This is a dark brown snake that, in life, showed no signs of a striped pattern but, as in other long­preserved adults, there is a faint bluish gray line on each side of the body on scale rows 3 and 4 anteriorly and 2 and 3 posteriorly, where a prominent pale lateral stripe would have appeared in a living specimen of Thamnophis eques eques , for example. The top and sides of the head are dark brown, and the upper labials are dark gray and virtually unmarked. The two lowermost rows of scales have light centers, but these markings are weak in comparison with several other preserved adults.

The anterior edge of each ventral is dark gray, almost black, and collectively is part of a long series of crosslines that show clearly through the overlapping part of the preceding scale. The unmarked posterior part of each scale is medium gray in coloration.

The chin, throat, and infralabials are pale yellow, but the throat is invaded by dark pigment on the two scales preceding the first ventral. There are small black markings on about half of the infralabials. The underside of the tail is also pale yellow, but the outer edges of the first 25 subcaudals are slightly marked with pale brown.

The anal plate is mostly pale yellow, but the dark pigment of the belly scales intrudes slightly on the right side of the body.

VARIATION AMONG THE PARATYPES: This subject is discussed in part under the diagnosis of obscurus, but there have been a few notable postmortem changes. Most conspicuous is the development of a row of light spots on each of the two lowermost rows of scales in all of the large­ and medium­sized snakes collected during the 1959–1965 period when we visited the Lago de Chapala. The edges of the scales in the two lower rows are dark, whereas the centers are pale, even white. The two light rows of spots are conspicuous in some specimens, less so in others. In the case of AMNH 91693 , the two rows of light spots barely showed in life (fig. 11). When examined in 2000, after being preserved in 1959, the light spots were prominent throughout the length of the body, especially on the lowermost row .

The two oldest specimens (AMNH 19544, 19545), preserved in 1919, have faded to dull gray. Most of their dorsal scales are edged with dark pigment, but the centers are pale. A faint suggestion of lateral stripes is gray rather than bluish.

Juveniles in life were pale to medium brown dorsally and grayish ventrally. Faint markings appeared dorsally in the form of light stripes near the head. After a long period of preservation most of them now show evidences of both striping and spotting. A particularly well­patterned juvenile female (AMNH 91694), collected at the town of Chapala on July 15, 1959, has faint middorsal and lateral stripes extending rearward to the tail. There are also small brown spots in two longitudinal rows occupying the areas between the faint stripes, and measuring about one scale in width and two scales in depth. There is also a smaller row of brown spots below each lateral stripe. The chin, throat, anal plate, and underside of the tail are unmarked yellow.

The coloration in life of three snakes from Jocotepec (AMNH 96842, 96844, 96845) was recorded, one soon after our return to base in 1965 and the others after they had been held in captivity for about a year. The following is a condensation of longer descriptions (capitalized color names are from Ridgway, 1912): Dorsal ground color olive. Lateral ground color Light Brownish Olive. Top of head olive. Labials Pistachio Yellow, sutures black. Belly Deep Bluish Gray­ Green. Lines across ventrals black. Underside of tail Olive­Buff. Chin and throat pale greenish yellow. Eye: Pupil black ringed by bright yellow, iris olive. Tongue: Dragon’sblood red, tips black.

Brumwell (1939) published a paper on variation in Thamnophis macrostemma (= T. eques ) which included material from Chapala, lake or town not stated. The specimens he examined were from the Edward H. Taylor–Hobart M. Smith collection preserved in earlier years, and his descriptions of them matched those of mine after my snakes had been preserved for decades. According to Brumwell, all juveniles were strongly patterned and the adults had developed pale lateral lines. Neither was true when my snakes were alive. What an advantage I had in being able to study the snakes in life and again after there had been strong postmortem changes.

REPRODUCTION: At each of the other lakes in Mexico’s transvolcanic belt where we collected series of specimens, gravid females were included. We kept them alive until the birth of their young, which were measured and weighed. No such females were found at the Lago de Chapala and thus no data are available on dates of parturition or number of young in obscurus. Several small specimens were obtained during July 1959 at the town of Chapala, some of them in storm sewers. The smallest was 244 mm in total length and another, with much of its tail missing, was 219+ mm. Based on measurements of newborn snakes from other lakes of the region, virtually all the juveniles we and others caught at Chapala were probably recent neonates. The largest specimen, a female also from Chapala, measured 1216 mm in length soon after it was preserved in 1959.

BEHAVIOR: We made pertinent observations on behavior, some of which did not come into focus until the end of July 1965, when we left the Lago de Chapala for the last time. Others, notably Allen G. Brown, had watched the snakes for hours and he permitted me to record many of his observations in my notebook. In the company of the late Charles M. Bogert, Brown and two other students spent a few weeks during 1959 doing fieldwork in the area. We joined them for a few days.

The town of Chapala is a resort on the lake roughly 40 km southeast of the large city of Guadalajara. A pier with benches and a ramada extended into the lake, and on weekends it was frequented by many people. Both observation and collecting were possible, the first by day, the second by night. I caught the holotype and two other eques by leaning over the edge of the pier at night and using my headlamp. The water was too low for me to reach it, but my wife, Isabelle, clung to my ankles, which gave me enough extra distance ‘‘to accomplish the impossible’’, as she recorded in her diary.

Brown (personal commun.) stated that it was normal to see the snakes swimming in the open water from the pier, sometimes far from shore, and even quite small ones moved about on the surface. He also said he watched several of the snakes swim toward the pier across wide stretches of open water. They would advance leisurely until they neared the pier, when they apparently detected motion and possible danger. They came right up to the pier at night. Brown and another student caught AMNH 82037, a female 654 mm in length, swimming far out in the open lake during daytime roughly midway between Chapala and the relatively distant Isla de Alacranes.

The available sample from the Lago de Chapala of 31 obscurus contains 11 males and 20 females ranging from neonates to large adults, including a heavy­bodied female 1216 mm (almost 4 feet) in length. This is a small number in view of the fact that we visited the lake during the summers of 1959, 1961, and 1964, and were in residence at Jocotepec at its western end for almost the full month of July in 1965. We rented a cottage for a headquarters from which we visited the Lagunas de Atotonilco and Cajititlán and the remnants of the Lago de Magdalena, and we collected almost all the way to the Pacific coast near Tepic. While we were in Jocotepec, I obtained much useful information about the behavior of Thamnophis eques obscurus , almost all of it negative. During the early evening of each of the many nights we were in residence I waded in the shallow water at the edge of the lake looking for snakes. The relatively small Thamnophis melanogaster was abundant and found during every foray, 11 in one evening. It was also available by day, chiefly hidden among rocks or under clumps of vegetation near the water’s edge. Thamnophis eques was conspicuous by its absence. I failed to obtain any of that species until the very end of the month when the level of the lake began to rise (about 15 cm in 24 hours on July 27–28). Meantime, a strong east wind drove water into areas that had remained dry all month. On the evenings of July 29–31 I found five obscurus, two of which were gorging themselves on small fishes that had ascended new channels into formerly dry territory. The following paragraph is condensed from my fieldnotes.

A pattern has emerged. While the lake was low melanogaster was conspicuous, eques totally absent. Apparently, melanogaster is adapted to a riparian habitat that it exploits year round. Contrariwise, eques remains quiescent during the dry season. I had explored the landward side near Jocotepec to where there were weed­choked fields through which a machete would have been necessary to penetrate. The four eques I found nearby were along a narrow ditch where many fishes, which had worked their way upstream when the wind and waves propelled water inland, were trapped when it receded. Does the eques population go through some form of estivation during dry weather? At least at the western end of the Lago de Chapala?

According to word received from Joseph Jordan of Jocotepec, our landlord for the cottage we occupied, Lake Chapala rose to a very high level after our departure on August 2, 1965. In a letter dated December 16, 1965, he admonished me, ‘‘When you make your next visit, do it in August or September, as the lake is real high and there are thousands of large snakes swimming wherever you look, and every little wash has several snakes in it. The lake, by the way, is at its highest level in twenty­one years, and my lower property is under five feet of water. It crossed the road in several places, but it is receding a little now.’’

The Jordans were a middle­aged couple and members of the large American colony in the towns of Ajijic, Chapala, and Jocotepec on the shore of the lake. They owned considerable property, including the weedchoked fields I could not penetrate. I had indoctrinated them soon after we first met in 1964 and had shown them several living examples of the two ‘‘water’’ snakes. I feel confident that Jordan’s ‘‘thousands’’ of large snakes were Thamnophis eques obscurus .

When the water was rising, Paul Ruthling collected two eques along the edges of flood­ ed fields near Jamay at the eastern end of the lake on August 19, 1919. These are AMNH 19544 and 19545, about the first of which Ruthling wrote (personal commun.), ‘‘It was very large and fat and 42 inches long.’’ Snakes approaching or exceeding a meter in length from the Lago de Chapala were all large in girth and heavy bodied, as indeed are large snakes from other lakes of the general region.

It is obvious that obscurus is a common snake in the Lago de Chapala despite the small size of our sample. Our experiences confirmed once again that ‘‘the collector must be in the right place at the right time.’’

A unique feature of the Lago de Chapala in 1959 was the presence of huge masses of water hyacinths of giant size covering the western half of the lake and completely halting all navigation. Manatees, Trichechus manatus , were introduced to eat the hyacinths and by 1965 the pestiferous plants were no longer a menace.

FOOD: The following is quoted from correspondence with Robert R. Miller: ‘‘The tiny fish trapped by receding waters of Lake Chapala include 12 juveniles of Goodea atripinnis , 2 young of Chapalichthys encaustus , one young of Alloophorus robustus , and 6 young of Poeciliopsis infans . The first 3 belong to the family Goodeidae , the last to the Poeciliidae .’’ I sent the fishes to him in October 1965. Probably any fish small enough to swallow would be eaten. No evidence is available about other food such as frogs, salamanders, and invertebrates. Captives fed on frogs.

WHAT OF THE FUTURE?: There have been many changes at the Lago de Chapala since we worked there in the 1960s. It is the principal water source for Guadalajara, Mexico’s second largest city. Water also is drawn for the production of electricity. Pollution is acute. León and Escalante (1993) stated that the inflow stream, the Río Lerma, ‘‘drains a basin of almost 52,500 km 2 receiving urban, agricultural, and industrial untreated waters.’’ The inflow to the shallow Lago de Chapala is highly polluted. Many native fishes have been locally extirpated and exotic species now constitute a large part of the catch. What effect are all these problems having on the population of Thamnophis eques obscurus ?

LAS LAGUNAS DE ATOTONILCO Y CAJITITLAN

Northwest of the Lago de Chapala are two small lakes, from each of which we obtained series of specimens of the Thamnophis eques complex that proved to be of exceptional interest because they are markedly different from those of the nearby big lake.

The Laguna de Atotonilco, highly irregular in shape, lies south of the settlement of Villa Corona, which has an elevation of 1350 m above sea level. The lake’s longest dimension, roughly northwest–southeast, is 13 km and its widest, toward the south, is about 6 km. Collecting both in 1964 and 1965 was exclusively at night at the north end of the lake. Fishermen cleaned their catches along the shore; flies and other insects congregated and they, in turn, attracted anurans. Some of the snakes were in shallow water actively swimming; others were secreted beneath an old boat. Using my headlamp, I stalked one, well in excess of a meter in length, until I was close enough to grab it even as it bit me savagely. It was the most continuously aggressive of the many T. eques I kept in captivity. The pH of the lake was 9 and the temperature of the water was 30° at 9:30 PM when tested in 1964; both figures were exceptionally high.

The Laguna de Cajititlán lies about 30 km almost due east of Atotonilco and its eastern extremity is 12 km north of Ajijic on the shore of the large Lago de Chapala. It has an east–west orientation and is 10 km in length, including a narrow neck of about 2 km at its western end. Its greatest width is approximately 4 km. During our visits in 1965 (July 15 and July 28) the part of the Laguna de Cajititlán adjacent to the village of the same name (elevation 1560 m) was clear and open, but its western end was a jungle of tules and smaller emergent vegetation. Boats of several sizes, including a large one with an outboard motor, were available. We tried several, but the prows of all were so high and sharply pointed that it was awkward to try standing with a long­handled dipnet hoping to lift a swimming snake. The ubiquitous boys who appear like magic whenever a gringo stops to do anything in Mexico and a lad whose father owned the powerboat came to our rescue. They stripped to their shorts, went overboard, grabbed snakes, and tossed them into my boat where I could catch them. We obtained 11 large Thamnophis eques , most from near the tules, some of which had broken horizontal reeds that were good perches for basking snakes. Another specimen was caught at night eating a small fish discarded by a fisherman, and identified as Chapalichthys encaustus by Robert R. Miller. The cloacal temperature was 24.2°; the water nearby was 24.7°. We made no attempt to record data for the snakes pursued by the boys and thrown into the boat.

We took samples of tules home with us. They were identified by Alfred E. Schuyler of the Academy of Natural Sciences of Philadelphia as Scirpus californicus which, despite its name, we were told is fairly common throughout northern and central Mexico.

The Cajititlán snakes matched those from Atotonilco in pattern and coloration. All had three longitudinal stripes. The middorsal one was wide and conspicuous. It involved three rows of scales, the vertebral and paravertebrals and, in a few cases, it even extended onto the edges of the adjacent scale rows so that the width of the dorsal stripe could be scored as 3+ instead of 3.

Clements (1963), based on his carefully researched ‘‘Pleistocene history of Lake Chapala’’ and the hills and mountains surrounding it for a considerable distance away, showed that during the Pleistocene Chapala ‘‘rose to a level almost 200 meters (656 feet) above the present lake level, spilling over into neighboring basins to the north and west and extending several kilometers farther into the state of Michoacan than now.’’ Clements also pointed out that presumably during the Wisconsin glacial age the water rose to ‘‘a level approximately 150 meters (492 feet) above the present lake level.’’ The high lev­ els of the Lago de Chapala in times past must have created vast habitats for aquatic and semiaquatic organisms.

During our month’s residence in 1965 at Jocotepec at the west end of the big lake, and after we learned there was a population of Thamnophis eques at Cajititlán, we became curious about whether that snake occurred elsewhere in the vicinity. Both our resident American and Mexican friends were skeptical. They said the area was very dry and not suitable for ‘‘water’’ snakes. We went exploring, however, chiefly driving along roads, and the vegetation of the region confirmed the semi­aridity. Someone mentioned a pond near Estipac west of the Laguna de Atotonilco. We drove to that settlement and, with the aid of a local resident, hiked ~ 2 km to a small pond where I caught a young snake (AMNH 94964). Subsequently, I found two adults at Acatlán de Juárez (AMNH 94924, 94925) along ditches through a small swampy area. After our return to home base, a review of the snakes collected by the Bogert party in 1959 revealed another taken approximately 8 km west of Ixtahuacán de los Membrillos, which is south of Cajititlán and only 6 or 7 km north of the Lago de Chapala, but separated from it by the Sierra el Travesano. That snake (AMNH 91692) is strongly striped (see fig. 13, lower) and very different from the stripeless ones inhabiting Lago de Chapala. These scattered localities suggested that the snakes of the Atotonilco–Cajititlán complex once ranged widely across a former large lake, as Tamayo (1949, II: 416–417) wrote (translation): ‘‘It is possible to suppose that the Lagunas Atotonilco and Cajititlán constitute the residue of an old large lake.’’

Douglas A. Rossman (personal commun.) reported that he and his students found another locality. They were driving southwest from Guadalajara on May 29, 1969. The highway passed over a pond on which they saw swimming snakes. Investigation revealed that Thamnophis eques was numerous. Seven eques and four T. melanogaster were collected and are now at Louisiana State University (LSUMZ). The environs of the small body of water were extremely dry and some of the snakes had taken shelter un­ der thorn bushes. The locality was the east­ ern tip of the Playa de Santa Cruz, an intermittent lake 5 km long that probably remained dry except after periods of heavy rain. There must have been a place, however, where the snakes could retreat (estivate?) during long dry periods.

In both 1964 and 1965, we attempted to explore some of the other small lakes south of Atotonilco, but we were blocked by deep mud at the Laguna San Marcos, and the Laguna Sayula was too saline at the time to support snakes and their food animals. The lower lakes apparently are gradually drying up. On a map published in 1973 (DETEN­ AL, Guadalajara sheet, 1:250,000), the Laguna Sayula is marked with hatching, indicating it is intermittent, containing water at times but dry at others. On a more recent map dated 1998 (INEGI, Guadalajara sheet, same scale), Sayula, San Marcos, and Zacoalco, which parallels San Marcos in part, are now all intermittent. Atotonilco and Cajititlán appear to remain much as we had found them. With only rains, a few springs, and small, mainly intermittent streams for increment, the entire area continues to desiccate.

Guerrero (1964), in his ‘‘Esquema geográfico de México’ ’, included a map showing the wet and dry zones of his country. Except for a small area adjacent to Nayarit, he designated all of Jalisco as semiarid.

Lago Cajititlán, a large and well­illustrat­ ed paperback written by Ramón Rubín and published in Guadalajara in 1960, reviewed the history of the lake in both Spanish and English. Rubín stated that the lake was 25 km due south of Guadalajara and that its surface occupied 16 million square meters. Following the Spanish Conquest, five small towns were established around the lake and remnants of their churches and other structures survived, some in excellent condition. In early times, fishing and agriculture along the lake’s perimeter were the principal activities. In the late 1940s, there was a strong movement in official Mexico to drain lakes to provide more farmland. Cajititlán was emptied, but trying to cultivate the slimy, muddy lake bottom was a dismal failure. Concurrently, there was a bitter battle to save the huge Lago de Chapala from a similar fate. In time, Cajititlán was refilled and re­ stocked, largely with exotic fishes. The snakes, presumably from small relict habitats nearby, eventually returned.

In all individual populations of Thamnophis eques summarized thus far the samples at hand were from separate lakes that are isolated in one way or another from other lakes of the transvolcanic belt. In the case of the Atotonilco–Cajititlán population, its range has been fragmented into two lakes and several isolated localities, all being remnants of a former large body of water that apparently is shrinking as aridity increases. For this unusual group of snakes with an unorthodox distribution I propose the name below.

Figures

Fig. 10. Map of the western end of the Lago de Chapala and several small satellite lakes, all in Jalisco. Delineated by W.H. Brandenburg.

Fig. 11. Upper. Adult female of Thamnophis eques obscurus (AMNH 91693), 834 mm in total length, from the Lago de Chapala, collected July 15, 1959, at the town of Chapala, Jalisco. Note the complete absence of pale longitudinal stripes. Lower. Venter of same snake. The anterior of each ventral scute is dark but not nearly as wide as in the snakes from the Cuenca de Michoacán.