472 PROCEEDINGS OF THE ACADEMY OF [April, 



Coluber has about the same range as Zamenis, but in the Nearctic 

 does not pass beyond the Rocky mountains. The distribution of 

 color most common in the genus consists of three rows of spots, but 

 in many species, especially in America, there is a disposition for these 

 to run together by their corners and form longitudinal stripes. This 

 is indicated more especially in some western and Central American 

 species, as C. subocularis and C. lineaticolUs, and culminates in C. quad- 

 rivittatus, which is spotted when young and wholly striped when 

 mature. C. ohsoletus confinis is in most respects generalized as to 

 pattern, and either this species, or less probably C. guttatus, both 

 chiefly Ocmulgian, are likely to be near the primitive form. 



In the Lower Sonoran Coluber has given off Arizona and Pihjophis, 

 the last having sent one species, P. melanoleucus , the most specialized 

 of the genus in the shape of its rostral, and in disappearance of the 

 head bands in adults, back into the Atlantic as far north as New Jersey. 



The genus Ophibolus, generally distributed in the Nearctic, except 

 in the northwest, is nearly related to Coronella of the western Palse- 

 arctic and Ethiopian. It does not seem possible to determine whether 

 the parent is represented by some color form of 0. gctulus or 0. doliatus, 

 but the coronelline affinity indicates with sufficient certainty that OpM- 

 holus must have come from a soiu-ce whose Palsearctic connections 

 were by way of the eastern or north Atlantic route. It may be 

 significant that while no subspecific distinction is warranted, most 

 Floridan examples of 0. gchdus getulus, which is the most vigorous and 

 extended species, seem to be rather generalized in the character of the 

 dorsal spots and bands, as between northern specimens of the same 

 form and the Louisianan 0. g. sayi. The further fact has a bearing, that 

 the specific characters of forms from the western part of the Nearctic 

 seem to show less stability than the eastern ones, especially so in 0. g. 

 boylii and 0. g. californice, thus pointing to a more recent development 

 of these subspecies. 



Liopeltis and Cyclophis are examples of the remarkable discontinuity 

 already noted in the range of certain other genera common to eastern 

 Asia and eastern North America. In the last Liopeltis covers the 

 whole Atlantic subregion, while Cyclophis is mamly Austroriparian. 

 Both extend to no great distance into the Sonoran. 



The case for Heterodon is not free from doubt. The form of its 

 rostral seems to suggest that it belongs among the types which have 

 been so freely developed in the dry Lower Sonoran, and its possible 

 relationship to the South American Lystrophis would accord with this 

 assignment, but on the other hand, the largest and most widely spread 



