Die BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
numerous finger-shaped, membranous, dark-colored prolongations 
of the skin, with a dense connective-tissue axis. In this part of the 
cutis he found some small insignificant blood vessels and also lymph- 
spaces, but neither nerves nor nerve-terminations. Therefore, in 
his opinion, these appendages are not to be considered as a sensory 
apparatus. Their function is unknown, and it also is unknown 
whether they are developed exclusively during the breeding season. 
That is all that we know about these remarkable structures, and 
I have only to add, that recently Nieden (:07, and :08, p. 659) has 
pointed out that the genus Trichobatrachus is identical with the 
genus Astylosternus, the latter name having been previously (1898) 
employed by F. Werner. 
No less than eleven specimens of this genus have been at my dis- 
posal, all coming from Kribe, Kamerun, and all belonging to the 
same species, Astylosternus robustus (Blegr.). 
From the study of these specimens I have established: First, that 
the hair-like appendages are present only in the males; they are 
wanting in all females. That agrees exactly with the second state- 
ment of Boulenger, but disagrees entirely with his first statement, that 
the character is more strongly developed in the female than in the 
male. As he based this strange assertion on the investigation of only 
two specimens, a male and a female, I feel sure that some mistake must 
have happened. I conclude therefore, that these appendages occur 
only in the males, and that they are certainly secondary sexual char- 
acters. 
Secondly, that these appendages do not attain the same degree of 
development in all male individuals, and that even in full grown males 
there are very conspicuous differences in this regard. 
In one jar there were preserved three adult males, apparently 
captured at the same time the largest being 115 mm. long, and all 
these large males showed very short appendages, in some places merely 
suggestions of appendages (Plate 4). Another jar contained two adult 
males of about the same size as the three, one specimen being, however, 
somewhat smaller than the three contained in the first jar, since it 
measured only about 100 mm. in length. The two larger specimens 
had a very remarkable appearance, caused by their very long hair- 
like villosities, which attained in the larger of the two (Plates 1-3) 
a length of about 20mm. In the same jar was also preserved a third, 
much smaller, male, measuring only 80 mm.; but this also was pro- 
vided with short, though quite distinctly developed appendages. 
Thus we see that in this species there are males with long and males 
