HORN EXPEDITION—AMPHIBIA. 
167 
The two specimens obt.ained during tlie dry season were of the dull colour 
hitherto described, but as in the case of Chiroleptcs platycephalus there is a very 
distinct brightening in colour iunnediately after the rainy season. In all specimens 
secured at this time the ground colour of the upper surface is yellow, with very 
light green mottlings. The toes and web are usually pink or warm-yellow. In the 
lightest ones the ground colour of the upper surface of the head and back is 
orange with green mottlings, on which are splotches of brown. 
The iris is deep orange, with dark brown spots. 
The ventral surface is a light yellow or yellowish-green. 
The whole body is puffed out, and becomes still more so if the animal be 
irritated. 
Along the middle of the body runs a light vertebral line, which in spirit 
specimens frequently turns pink in colour. This pink colouration of certain parts 
under similar circumstances has been already noticed by Mr. Fletcher in the case 
of L. sahninii, and is true also of L. ornatus, in which the piidcish tinge pi-esent in 
the living form on small tubercles may spread to adjacent parts when the animal 
has been in spirit. 
The webbing varies considerably. Mr. Fletcher, in describing specimens from 
the Emu Plains, New South Wales, noted that the toes were fully webbed, whereas 
in the British Museum Catalogue they are described as two-thirds webbed. 
In my specimens there is a certain amount of variation which does not seem 
to depend upon the sex, or even upon the maturity of the specimen. They vary, 
males and females, mature and immature ones, from two-thirds to fully webbed. 
The webbing usually extends to the tip of the first, second, third, and fifth toes, 
and in the most webbed ones a fringe runs right to the tip of the fourth toe. 
The question of webbing has been previously referred to by Mr. Fletcher in 
dealing with other species.* Judging from the examples of Z. ornatus and H. 
picius, it is somewhat diflficult to determine how far webbing may be relied upon 
even as a specific character; as a generic character it may probalily be regarded as 
of very little value indeed. Within the limits of the genus Chiroleptes we have 
one species, C. platycephalus, which is normally fully webbed, and another, C. 
brevipahnatus, which is not more than one-third webbed. In the case of Limno- 
dynastes, again, we have a genus the species of which are as a general rule but 
* Vol. iii., Proc. Linn. Soc. N.S.W., 1S92, p. 18. 
