478 The American Naturalist. 
non-venomous snakes is well known. That authors are not ag 
to the principles on which these animals should be classified 
learned by comparison of the systems published by Duméril and 
in 1853, Giinther in 1858, myself in 1886 (Proceeds. Amer. 
Society), and Boulenger in 1892 (Reptilia of Zoology of British 
It had appeared evident to me that a further examination of the a j 
omy of these reptiles is necessary before a correct account ot i 
mutual relations can be given, and that the organs of the reproducti 
system especially, were likely to yield important results. W rt 
object of obtaining light on this question, I have made an exami a 
of the hemipenis, and have obtained valuable indications of relationsh 
which have been hitherto unknown. I have had the opport 
examining material from the collections of the Academy of Phila 
of the U. S. National Museum, and of my own cabinet. I present 
a general synopsis of the results, reserving for a future occas 
publication of a fuller illustrated memoir on the subject. The 
classification is more in harmony with the systematic indi 
obtained from the study of other vertebrata, than those hitherto a 
which is itself an indication of its greater approximation to 
Some points remain obscure, and many details are omitted fi 
present prodromus. 
The hemipenis of the Ophidia is traversed by a groove which 
the muscular investment to the internal integument (or external 
ument when the organ is retracted), which commences at t 
internally, and soon turns to the external side of the organ al 
tinues to its extremity. This is the sulcus spermaticus. This 
always bifurcated in venomous snakes, and I find it to be equi 
cated in many harmless snakes. The investing tissues may 
correspond with this bifurcation. Thus the hemipenis may 
nay bifureate. we states that it is bifurcate in venomous 
ut it is not so in Hydrophis hardwickii, B us semifasciatus, 
cephalus coronatus, ete., while it is bifureate in many masm 
Next to the bifurcation of the sulcus in importance, is the nature 
surface of the external investment (internal when retracted). 4 
most perfect types both venomous and non-venomous, this 
a suctorial function. Their borders are generally 
i s0 deeply divided into papille as to lose their orig 
These papillæ may be the seat of osseous s 
bristles or spines, which become larger toward the middie oft 
and lose their mutual membranous connections. These isol 
