genus, the O. triangulum and 0. coccineus, have always been regarded as 

 distinct species^ and so numerous are their differential characters, in col- 

 oration, size, and squamation, that this view would seem to rest on a sat- 

 isfactory foondation. I find, however, that individuals exist which rep- 

 resent every stage of development of each character which distinguishes 

 them, although certain types appear to be more abundant than the interme- 

 diate ones. 0. triangulum is a species of larger size, with two temporal 

 plates, a row of large dorsal spots, and other smaller ones on the sides, on a 

 grayish ground ; with a chevron, and often other marks on the top of the 

 head, and a band posterior to the eye. 0. coccinemis a small snake with a 

 small loreal plate and one temporal shield ; color red, with pairs of black 

 rings extending round the body, and no markings on the head excepting 

 that the anterior ring of the anterior pair crosses the posterior edge of 

 the occipital shields, forming a half collar. The transition is accom- 

 plished thus : The lateral borders of the dorsal spots of 0. triangulum 

 break up, and the lateral spots become attached to their anterior and 

 posterior dark borders. The chevron of vhe top of the head first breaks 

 into spots, and then its posterior portions unite with each other. The 

 borders of the old dorsal spots continue to the abdomen, where the 

 remaining lateral portions finally meet on the middle line, forming a 

 black line. This breaks up and disapijears, leaving the annuli open ; 

 and these are then completed in many specimens. The general colors 

 become more brilliant and the size smaller. The head is more depressed ; 

 in immediate relation to this form, the loreal plate is reduced in size, 

 and the two temporal shields of 0. triangulum are reduced to one. Every 

 form of combination of these characters can be found, which represent 

 six species of the books (in North America), viz : 0. triangulum^ O. doU- 

 atuSf 0. annulatnsj 0. gentilis, 0. amaurus, and 0. coccineus. The oldest 

 name is the 0. doliatus, Linn. Another series of specimens resemble 

 very closely those of the subspecies coccineus ; in fact, are identical with 

 them in color. The loreal shield is, however, extinguished, and the rows 

 of scales are reduced by one on each side. These specimens simply carry 

 one degree further the modifications already described. Yet, on account 

 of the constancy of these characters, I am compelled to regard these 

 individuals not only as a distinct species, but, on account of the absence 

 of the loreal plate, as belonging to another genus. This is the Calama- 

 ria elapsoidea of Hclbrook ; the Osceola elapsoidea of Baird and Girard. 

 It affords an illustration of the principle, which I have elsewhere insisted 

 on, " that adjacent species of allied genera may be more alike than remote 



I 



