56 
Annals of the Transvaal Museum. 
with large white blotches. Lips crossed by two or three oblique dark 
bands on each side. 
Length from snout to vent 20 mm. 
We strongly suspect that this species belongs to the same genus as the 
frog described and figured by Bottger under the name of Hemimccntis 
horrid a ( 2 , p. 282, and 6 , p. 492, Taf. Ill, fig. 14), and it is quite possible 
that the two are even specifically identical. According to Bottger’s figure 
of the hand (l.c. PI. Ill, fig. 14c) the second huger reaches further than the 
fourth, but in our species as already stated the second does not reach as 
far as the fourth ; further the shape of the snout in the two species differs. 
If we are correct that Hemimantis (Arthroleptis) horrida belongs to the 
same genus as Microphryne it is of interest to note that the endemic 
Ranidae of Madagascar with one single exception ( liana labrosa) are all 
provided with supernumerary phalanges." 
Type, No. 1155, in the Transvaal Museum. 
Genus Rappia, Gtlir. 
R. RUTENBERGI, Bttgr. (6, p. 510). Two examples, evidently 
juvenile, from the eastern region. 
These specimens agree with Bottger’s description (6, p. 510), but 
differ in the two following respects : the tympanum is more or less 
distinct, and the toes are about three-quarters webbed. 
The larger specimen measures 16 * 5 mm. from snout to vent. 
It is from consideration of the colour pattern mainly that we have 
referred our specimens to this species. 1156-1157. 
Genus Meg ALIX ALUS, Gtlir. 
M. MADAGASCARJENSIS, D.B. Forty-three examples from the eastern 
region, and one from the plateau ; localities include Ambilo, Folohy, 
Analamazotra, Ambatoharanana, Ambobidratrimo, and Tananarive. 
In many of our specimens the snout is subacuminate rather than 
rounded, and in this respect some of the specimens would appear to agree 
better with M. tricolor , Bttgr. rather than with M. madagascariensis . 
Moreover the prevalence of bright yellow colouration would also appear to 
suggest the same conclusion (vide 13 , key to the genus, p. 65). We may 
point out, however, that no true conception of the original colour can be 
obtained by the examination of spirit specimens : thus though Mocquard 
(l.c.) describes for M. madagascariensis “Face dorsal hnement ponctuee 
de noir sur fond grisatre. Face ventral blanc sale,” the colours of this 
species in life may be very different. 
We have placed our specimens under M. madagascariensis for the 
reason that a small external metatarsal tubercle, said to be absent in 
M. tricolor , is present in our examples. 
We note the following characters in our series. The tympanum may 
be hidden or distinct. The belly may be smooth or may have glandular 
granulations, or the belly and thighs may be distinctly granular. We find 
that this granulation of the belly and thighs is a very variable character 
(the granulation of these parts is often combined with the subacuminate 
* We are not including liana mascareniensis , as this species is very widely distributed 
in Africa and belongs to a group of the genus Rana which is characteristic of Africa ; it has 
no allies in Madagascar, and is in all probability a comparatively recent immigrant to that 
island. 
