Lacerta. 31 



3. It is not necessary to give reasons for regarding tlie parietal 

 foramen (in relation to the vestige of the pineal eye) as a primitive 

 character. It is absent in three species only of the genus Lacerta — L. 

 jacksonii, vatiereselli, and echinata ; also in a few African species of 

 other genera.* 



4. Considerations derived from the study of other families of 

 Lizards lead me to assume that the original condition of the nostril is 

 to be pierced in the centre of a single nasal shield. This condition is 

 not found in any of the Lacertidse. The next step is for the nasal to 

 divide into two — nasal proper and post-nasal — without the first labial 

 entering the nostril, as we find in some species of Nucras and, as an 

 exceptional occurrence, in Lacerta agilis and L. parva. 



As a further stage of evolution in the genus Lacerta, I regard the 

 participation of the rostral to the border of the nostril, as in L. viridis 

 and ocellata, and more or less constantly in some of the members of 

 the L. muralis group. A step further still, the nasal portion of the 

 rostral becomes detached as a small shield in front of the nostril, as in 

 some specimens of L. danfordii, which in this respect shows the 

 farthest departure from the primitive type in the genus Lacerta. In 

 rare cases {L. agilis, L. vivipara) it is the lower part of the nasal 

 which has been severed to form a distinct shield between the rostral 

 and the nostril. 



In L. agilis,' as in L. vivipara, the superciliaries and the supra- 

 oculars are in contact with each other ; only in rare cases are a few 

 granules intercalated between them. The absence of granules is 

 retained in many specimens of typical L. viridis, in L. peloponnesiaca, 

 and, as an exception, in L. taurica, ; in other species a more or less 

 complete series of granules separates the superciliaries from the 

 supraoculars, and in the extreme form L. oxycephala the series is 

 often partly double, whilst the first supraocular shows a tendencj'^ to 

 break up into small scales. In some species of allied genera, such as 

 Acanthodactylus and Eremias, the disintegration of the borders of 

 the supraoculars into small scales or granules may be carried further 

 still. 



In L. agilis, the temple usually bears very large shields, few in 

 number, but there is much variation ; in L. parva a finer scaling is 

 the rule, and as we proceed in the L. muralis series, we reach a stage 

 in which the temple is covered with minute granules, with or without 

 a central (masseteric) shield, which can be traced back to L. agilis. 

 The large upper temporal shields and the tympanic shield may also 



* Cf. Degen, t. c. p. 30. I have previously used the presence or absence of 

 the foramen for distinguishing species in the genus Draco (P. Z. S. 1897, p. 198). 



