Geophis, AND OTHER

Savage, Jay M. & Watling, James I., 2008, Not so rare snakes: a revision of the Geophis sieboldi group (Colubridae: Dipsadinae) in lower Central America and Colombia, Zoological Journal of the Linnean Society 153 (3), pp. 561-599 : 596-598

PROVENANCE OF GEOPHIS AND OTHER

SNAKES INCORRECTLY CATALOGED FROM ‘ PANAMA SABANAS’

Myers (2003) briefly reviewed the listing in the catalogue of the ANSP for Geophis brachycephalus (ANSP 24723–24734), Geophis godmani (ANSP 24722), and Trimetopon slevini (ANSP 24717–24719) from ‘ Panama Sabanas’. As the three species are all upland forms that co-occur in extreme western Panama and all other ‘Sabanas’ material is from lowland central Panama, Myers concluded that their association with that ‘Sabanas’ locality must have been a cataloguing error and that the Academy specimens were part of the Panama Snake Census (PSC) collection from Panama: Chiriquí: Finca Lérida (above Alto Quiel). Here we provide further details from Dunn’s notes regarding the provenance of the specimens in question.

Both the ‘ Panama Sabanas’ and the Finca Lérida series are part of Herbert C. Clark’s PSC collections. Clark, then Director of the Gorgas Memorial Laboratory, in Panama City, instituted a survey of Panama snakes beginning in 1929 and continuing through 1953 (see Dunn, 1947, 1949; Myers, 2003). The intent was to assess the relative medical significance of venomous snakes. Bounties were paid for specimens, usually heads, and specimens accumulated at various centres throughout Panama.

In 1949, Emmett Reid Dunn published his classic paper on the distribution and abundance of Panama snakes based primarily on the PSC collections (10 693 specimens) and 1073 snakes from natural history museums. The PSC specimens analysed in that report were from west-central and eastern Panama and grouped by general locality for this purpose as: Coclé- Herrera, Sabanas, Chagres, and Darién. The Sabanas material (3914 snakes) was from a variety of Pacific lowland localities, both east and west of the Panama Canal, between Panama: Panama: Cañita and Capria, respectively. Many of these have specific locality data but others are catalogued from more inclusive sites (e.g. Chorrera to Arraiijan) or simply Panama Sabanas (usually Sabanas in Dunn’s notes). These localities all lie in an area having a long dry season (mid-December to April) that originally supported scrub, shrub or savanna vegetation but has been anthropogenically modified to an extreme beginning in preconquest times (Bennett, 1968).

The Lérida PSC collection was not included in Dunn’s (1949) analysis but all Sabanas snakes collected prior to 1946 were. Significantly, Geophis brachycephalus , Geophis godmani and Trimetopon are not referred to anywhere in the 1949 paper. However, Dunn (1947) had published an often overlooked report prescient of the 1949 paper on the snakes of Lérida some 2 years earlier based on PSC material (533 specimens) and seven others in museum collections. This series included 255 Geophis brachycephalus , three Geophis godmani and 26 Trimetopon slevini. The PSC material from Lerida had been accumulated by Tolef B. Mönniche, the owner of the finca, beginning in 1940.

A review of Dunn’s notes accumulated over many years, at least beginning in 1928, and organized as a preliminary synopsis of the ophiofauna of lower Central America, clarifies the source of the Sabanas specimens of questionable locality data. Part of Dunn’s entries for G. brachycephalus list the following:

‘Lerida cl 1940 17 + ERD 1

1941 17

1942 35

1943 99 L tot 1947 = 255 1944 45’

In this notation, as throughout Dunn’s notes, cl (= Clark) is used for snakes from the PSC. Year and number of specimens are indicated through 1945. The total of examples from Lérida (L.) through 1947 is also noted and is the same total as listed in Dunn’s 1947 paper. No G. brachycephalus from Sabanas or Panama Sabanas are listed among the other 69 cited from Panama. The number is actually 65 as three snakes from Panama: Chiriquí: Boquete, in an extensive series (63 snakes) collected by Joseph R. Slevin (1942) (sl in Dunn’s notes) represent other species of Geophis (Downs, 1967) and include the holotype of G. championi (BMNH 1946.1.1.77). It seems clear from the above evidence and confirmed below that ANSP 2423–2434 catalogued as being from ‘ Panama Sabanas’ are also from Lérida.

Recall that most snakes from the PSC are heads or heads with a short segment of the body and after identification, unfortunately, Dunn did not keep and/or deposit all of them in museums (see Myers, 2003). He did, however, retain rarer forms and smaller complete specimens. The ANSP contains the following G. brachycephalus catalogued from Lérida: ANSP 21699, a specimen collected by Dunn in 1939, and the following PSC specimens: 22922–22938, ANSP 23877–23879, ANSP 24766–24769.

The situation for Geophis godmani is even more clear-cut. The pertinent material in Dunn’s notes is as follows:

‘Lerida cl 1940 1h, 1941 1h, 1943 1h.’

He makes no mention of other specimens from elsewhere in Panama. The ANSP collections contained only three heads (h in Dunn’s notes) all from the PSC, two from Lérida: ANSP 22910, ANSP 24765, and the putative Sabanas example (ANSP 24722), now lost. Obviously ANSP 24722 is one of the three heads recorded from Lérida by Dunn. Note that typographical errors list ANSP ‘24756’ for one Lérida snake (Downs, 1967) and ANSP ‘24723’ for the ‘Sabanas’ example (Myers, 2003).

The pertinent Dunn notation for Trimetopon slevini is:

‘Lerida ANS (2) cl. 19411 - 19427 19439

The notation 19439 apparently refers to the number of specimens collected by the PSC from 1940 to 1943. ANS (2) apparently refers to the two paratypes ( ANSP 21700–21701 ) collected in 1939 prior to when the PSC began obtaining material from the finca or to ANSP 22939–22940 . The ANSP catalogue lists four other PSC Lérida specimens: ANSP 23875 , ANSP 24770–24772 and three specimens of T. slevini ( ANSP 24717–24719 ) are catalogued from ‘ Panama Sabanas’ for a total of 11 Academy specimens of the species. Clearly , the ‘ Sabanas’ specimens are part of the nine collected for the PSC from Lérida. The only other specimen of this species listed in Dunn’s notes is the holotype ( CAS 78938) from Panama: Chiriquí: near Boquete .

Although not noted by previous authors, another upland snake, Ninia psephota , is represented in the Academy collections by a PSC specimen (ANSP 24720) purportedly from ‘ Panama Sabanas’. Dunn’s notes on N. psephota state:

‘Lerida cl. 1940 31941 31942 13 1944 4 [a line has been drawn through these] Lerida 38’

No mention is made of any members of this species from elsewhere in Panama, except the series of 61 snakes from nearby Boquete reported on by Slevin (1942). The number of snakes of this species listed from Lérida by Dunn (1947) is also 38.

A single specimen of Liophis epinephelus (ANSP 24721) listed in the Academy catalogue as being from ‘ Panama Sabanas’ has not been mentioned by previous authors. A typed label in the jar containing this specimen reads ‘Finca Lerida, Chiriqui, PSC’. Dunn’s notes record 52 (57 in Dunn, 1947) specimens of this species from Lérida and 19 heads from Sabanas. As the Academy specimen is complete and has associated locality data, it must have been collected at Lérida. In an obvious lapsus Dunn (1947) lists the number of specimens of this form from Lérida as 31, duplicating the number recorded for Lampropeltis triangulum.

Significantly, the catalogue numbers for all the reputed ‘ Panama Sabanas’ PSC specimens that obviously came from Lérida are in a single sequence (ANSP 24717–24734). The previous series, ANSP 24703–24716, are PSC snakes recorded in Dunn’s notes and the Academy catalogue as from ‘Sabanas’. We can only conclude that there was a cataloguing error where Panama Sabanas was entered as the locality for the next following 18 specimens actually collected from Lérida.