POISONOUS SNAKES OF NORTH AMERICA. 415 



its food is almost wholly made up of mice aud other rodents, and he 

 consequently considers it decidedly useful, aside from its venomous 

 qualities. It seems hardly advisable, however, to suggest protection 

 for this species on this account, but I would advise that the farmers 

 spare the life of every large harmless snake on their land, and there 

 Avould be no harm in killing off every Eattler, for harmless snakes will 

 destroy the mice fully as well as the poisonous ones. 



The Gulf-coast Massasauga. 



Sistriinis cafenaiuK coiisors," (B. &G.)- 



1853. — Crotalophonis co)i.sors, Baird and Girard, Cat. N.Ara. Serp., p. 12.— Du- 

 MERiL and BiBRoN, Erpet. G6ner., vii, ii, p. 1482 (1854).— Baird, Pac. R. 

 R. Rep., X, Reptiles, p. 14 (1859). 



1883.— Sistruriis caieiudxs, var. cotimrs, Garmax, Rept. Batr. N. Am., I, Ophid., 

 p. 176. 



1892. — ? Sistrunts catetiatiifi, Garman, Bull. Essex lust., xxiv, p. 4. 



Fiffure.—BAmu, Pac. R. R. Rep., x, Rept. pi. xxiv, fig. 8 (1859). 

 The status of the present form is very doubtful. It was described by 

 Baird and Girard from a single specimen collected at Indianola, Tex., 

 which now appears to be lost. The original description does not fur- 

 nish any very tangible character by which to separate it from typical 

 S. catenatus with 25 scale rows, but its scutellation is compared chietly 

 with 8. miliaritis, which seems to indicate that it may have had the pre- 

 ocular and posterior nasal separated, although the otherwise so char- 

 acteristic color pattern of the head is that of S. catenafns. The figure 

 in the Pacific Eailroad Report [pi. xxiv, fig. 8], gives only the top of the 

 head, but the above suggestion is strengthened by that figure, which 

 certainly seems to show a separation of the shields mentioned by a loreal, 

 but whether by an upper loreal, detached from the anterior portion of the 

 preocular, or by a large loreal jjroiier, is not clear, although the former 

 alternative seems most prol)able. In that case we have probably to do 

 with an individual variation only, and the only ground for the sei)ara- 

 tion of the subspecies would be the smallness of the dorsal spots. The 

 25 scale rows would then distinguish it from the subspecies S. c. edivardsii. 

 lam inclined to think that the *S'. catniatus reported by Garman (Bull. 

 Essex. Inst., xxiv, 1892, p. 4.) from Deming's Bridge, Matagorda County, 

 Tex., not very far from the type locality of IS. conmrs, and which, like 

 the latter, had 25 scale rows, belong here. They have, moreover, 48 to 

 51 dorsal blotches. 



For the sake of completeness, I add the original description by C. 

 Girard. 



Description. — Twenty-five rows of dorsal scales, all carinated except 

 the two first rows on either side. Seven series of blotches, one dorsal 

 and o on each side, all very small. A yellowish white line passing from 

 behind the nostril below and behind the eye. 



* From the Latin consors, a partner, companion, or relative. 



