ALLIED TO LACERTA MURALIS. 3 



Affinities iniccrtaiii : 1j. hrnndti De Fil., jncksoni Blgr., vaiiereselli Torn., jayakan 

 Blgr., princeps Blanf. 



I have no doubt the authors referred to would now repudiate many of these su^es- 

 tions, but it is instructive to recall them with the object of showing how opinions 

 differ, and how much has to be done in order to arrive at a more correct understanding 

 of the relationships of tliese Lizards. 



The arrangement I would propose, after a careful consideration of all the characters, 

 is as follows : — 



Section I. {Lacerta, s. str.). L. arjilis L., parva Blgr., viridis Laur., princeps Blanf., 

 ocellata Daud. 



Section II. (Grt^/o^/«, n. n.), L. ntlantica Peters & Doria, ccesaris Lehrs, qalloti 

 1). & B., stehlini Schenkel, siviovi/i Stdr. 



Section III. [Zootoca Wagl.). L. v/vipara Jacq., fraasii Lehrs, derjuginil Nik., 

 praticola Eversm., vaiiereselli Torn. 



Section IV. {Podarcis Wagl.). L. tanrica Pall., peloponnesiaca Bibr., muralis Laur., 

 chlorognster B\gi:,jacksonii Blgr., hrandtii Ue Fil., Icevis Gray, jayakari Blgr., 

 danfordii Gthr., oxycephala U. & B., mosorensis Kolomb., dugesii M.-Edw. 



Section V. (Centromastix, n. n.). L. eehinata Cope. 



Section VI. {Thetia Gray). L. perspicillata D. & B. 



The first section corresponds exactly to Group I. of Bedriaga and Group IV. of 

 AVerner ; the inter-relationships between its components have been recognised by all 

 recent authors, and are so obvious that no doubt can be entertained as to the species 

 included constituting a natural association, although, in the days of Dumeril and 

 Bibron, Lacerta ocellata was placed with L. muralis in a different section on account 

 of the granular dorsal scales. But the group is hardly capable of definition, so 

 closely does it merge into tlie three following. 



The second section, confined to the Canary Islands, I have no doubt was derived 

 from some now extinct form closely related to L. ocellata, var. pater, inhabitiiiij 

 N.W. Africa, exceptional specimens of which still show traces of li^ht longitudinal 

 streaks ; the two postnasals of L. ocellata have been reduced to one. 



The third section, which I regard as directly derived from L. aqilis, has been 

 recognised by Bedriaga and Werner; it may be defined as combining a short snout 

 (usually shorter than the postorbital part of the head), a single postnasal, lar^-e scales 

 (25-50 across the body, 14-22 on the gular region), a strongly serrated collar, few 

 femoral pores (5-15 on each side), ventral plates as in Section I., and no teeth on 

 the palate. It should be kept distinct from the L. muralis association, altliough 

 approaching very closely L. muralis, vars. breviceps and caucasica. 



The fourth section is so completely linked with the first through L. agilis, parva. 



