t?r)2 ^Ir. G. A. Bonlenger on the Races and 



America. Close to R. cateshiana, lliore is another Nortli 

 American species, R. septenfrionu/is, in my opinion derived 

 trom it, in wliicli the fold is either present or ahsent, according 

 to individuals, but when present is short and very flat, with 

 the spots and niarblings irregularly distributed over the body. 

 Such a type leads to the state of things in R. esculenta, 

 vnr. ridibunda. 



Bolkay alludes to the transverse expansion of the dark 

 spots on the back as an in)portaiit character of R. c/iinensis, 

 but such transverse markings are by no means the rule in 

 this variety, some specimens of which are, on the contrary', 

 longitudinally streaked, as is often the case in the typical 

 form and the var. letsoyive, but never in the vars. ridi' 

 bunda and saharica. I may here mention that specimens 

 with the black markings forming cross-bars on the back 

 are exceptionally met with, not oidy in the var. ridibunda^ 

 but also in the typical form (females from Rivoli and 

 Verona). 



The light vertebral streak or band is very frequent in the 

 typical form and the vm?,. lessonce and cZf/«e?m*, less s ) in 

 the var. ridibunda, in which it is generally broader, and 

 usually absent in the var. saharica. I do not think this light 

 vertebral streak, which occurs in so many species, is to be 

 looked upon as a primitive character ; the frequent cases ui 

 deviation of its course from the straight line (most strongly 

 marked in specimens from Calcinaro and Cadillac) suggest 

 a different interpretation, and, in the present state of our 

 knowledge, its signification is highly problematic, as is that 

 of a light line along the inner side of the up|)er surface of 

 the leg which, in many Oriental and African species, often 

 acconi panics the vertebral streak, and exceptionally occurs 

 in R. esculenta, var. chinensis ( ^ from Japan). Both streaks 

 are absent in all American species, with the single excep- 

 tion of R. cantabrigensis, Baird, the representative of the 

 European R. arvalis, Nilss. 



The Skull. 



The osteological characters appealed to by Bolkay are 

 evidently derived from an examination of a very small 

 number of specimens ; put to the test of a larger material 

 they j)rove to be worthless for defining sj)ecies, 



I am especially surprised at his st;iteinents concerning the 

 nasal and fronto-parietal bones. Although usually in con- 

 tact with each other in full-grown specimens of the typical 

 form, as described and figured by Ecker, Fatio, and otliers, 



