196 
G. J. FEJÉRVÁRY 
praeocular (the more or less accentuated shields forming the conti¬ 
nuation of those on the under border of lower eyelid are not taken into 
consideration). Before the suboculare generally four (rarely 3) s u p r a- 
labials; behind the subocular generally 3, (rarely 2 or 4) supralabials. 
Temporal region covered by small, tolerably equal-sized, polygonal, 
slightly convex shields, the smallest of which form the series round the 
well developed 1 masseteric shield, latter sometimes split in two 
or more parts, yet present in all individuals examined; 
it is separated by 1—3 rows 2 of temporal shields from the supratemporalia, 
rarely meeting with them directely in a suture of varying length; in the 
examined specimens the massetericum is separated by 2—5 scale-rows 
from the tympanale, by 1—4 rows from the supralabialia and by 2—4 rows 
from the postocularia. Five to eight supratemporal shields; 
first supratemporal 3 longest, or at least generally broadest (highest) 
amongst other supratemporalia; these greatly vary in size, even on right 
and left side of the same specimen ; they sometimes diminish in size 
towards the back or appear tolerably equal sized or again alternate in circum¬ 
ference ; the first one’s length generally exceeding its height. Tym¬ 
panale usually strongly developed, at times split yet present in all 
specimens. (PI. I. fig. 3.) 
54—68 scales across middle of trunk. 4 Scales of trunk (PI. 
I. fig. 5.) smooth, or but very faintly keeled at the top, 5 (Coruna specimens) 
rather small sized, moderately convex, generally rather rounded (slightly 
rhomboidal in several places of trunk), sometimes oval (in several places 
of trunk slightly elongate-hexagonal) ; in both sexes these scales increase 
in size towards the flanks, flattening and acquiring a more rhomboidal 
shape. Three to four rows of flank-scales correspond to the outer margin 
of one ventral. Scales on upper surface of tail more-or less distinctly 
keeled ; their hind border very variable in shape ; widely rounded, somewhat 
pointed and nearly straight, or more or less distinctly pointed. No median 
row of larger scales ; larger ones are however here and there present on the 
median line of the tail, but these do not form a continual row T , and cannot 
be compared to the well developed median scale rows on the tail of the 
1 Excepting the Lisbon semiad. Ç and the Coruna ad. $ which present very weakly 
developed masseteric schields. 
2 Counted always in a straight line, and if configuration of scale-rows is not equal 
in respect to number, the smaller one is taken. 
3 All shields, large and smaller, bordering the parietal until its hind border are 
reckoned as supratemporalia. 
4 The supplementary rows of ventralia are here included, ventralia not reckoned.. 
5 PI. I. fig. 56. — According to JBoulenger (4, p. 362. ) «the dorsal scales are some¬ 
times distinctly keeled». 
