NOVITATES ZooLoaic.vE XL. lil.'iO. 115 
DR. KARL JORDAN'S EXPEDITION TO SOUTH-WEST AFRICA 
AND ANGOLA : HERPETOLOGICAL COLLECTIONS. 
By H. W. PARKER, B.A. 
Department of Zoology, British Museum (Nat. Hist.) 
(With two text-figures.) 
■jVrO large area of the African continent now remains completely unexplored, 
-'■ ' and in many regions the herpetological fauna is remarkably well known. 
But during recent years there has been a tendency towards an intensive studj' of 
some of the more easily accessible regions, so that the growth of our knowledge 
has not been uniform. For this reason alone the jiresent collections would have 
been of considerable value, but they have an additional and greater value owing 
to the fact that the areas selected for field-work were chosen to include as many 
different types of terrain as possible ; full details of these localities, together with 
details of their topography, climate and flora, will be found in Dr. Jordan's intro- 
ductory article. The author wishes to express his indebtedness to Dr. Jordan, 
not only for the privilege of studying the collections, but for much valuable 
assistance and information concerning habitats and habits ; acknowledgments 
are also due, and made with gratitude, to Mr. A. Loveridge for much enlightening 
correspondence and to M. Gaston de Witte, who very generously placed the 
magnificent collection of the Congo Museum at the author's disposal. 
Dr. Jordan's collections contain over 700 specimens, representing 95 species 
and subspecies, of which six are believed to be new to science. Their study has 
resulted in the discovery of numerous points of systematic interest which are 
embodied in the following notes, but some matters of more general interest 
concerning distribution and zoogeography may conveniently be discussed first. 
The bulk of the collections were made in strongly contrasting regions. In 
Angola most of the material (40 species) was collected in the heavily forested areas 
about Congulu and Quirimbo, whereas in S.W. Africa, though not geographically 
far distant, most of the species were taken in dry regions of granite, gneiss and 
sand. But, in addition, collections were made in open forest country both in 
Angola and in Daniaraland, as well as in dry granitic and sandy areas in Angola 
analogous with those in S.W. Africa. Analysis of the lists of species taken in 
these different climatic and vegetational zones emphasizes the enormous effect 
which these environmental factors play in determining the composition of the 
fauna, and the facts may be summarized for each zone as follows : 
A. ForeM and Swamp in Angola (Congulu and Quirimbo). 
Of the 40 species collected in these localities only 4 were collected in the 
dry zone of Damaraland ; three of these are widespread, tolerant species : 
Boaedon lin^ahui, Oerrhonaurus flavigularis and Biifo regidaria, and the fourth, 
Agama planiceps (q.v. infra), shows a distinct tendency towards subspccilic 
difierentiation under the different conditions. The facies of the remainder of 
the fauna is distinctly that of the equatorial Rain Forests, for 64 per cent, of the 
species are strictly confined to that area or to the outlying forest islands. Many 
of these forms are to be found widely distributed through the whole of this zone, 
116 NOVITATES ZOOLOGICAE XL. 1936. 
but a considerable proportion are essentially species of the Congo Basin and are 
not known to occur in the Cameroon-Gaboon area or the western forest province. 
The remaining species are either apparently indigenous in Angola (17 per cent.), 
widespread species such as Causus rJwmbeatus, which was not actually taken in 
Damaraland but undoubtedly occurs there, or species widely distributed in the 
Savannah countries bordering the Rain Forest and sometimes encroaching upon it 
(17 per cent.). Some of the indigenous species, such as Rana albolabris acutirostris 
and Leptopelis jordani, are obviously closely allied to, and are probably derived 
from, true Rain Forest species. It is thus apparent that, as might be expected, the 
forests of western Angola are essentially simOar faunistically to the forests of the 
Congo basin, but that some slight degree of differentiation has taken place, giving 
rise to new, indigenous species ; and further, their position on the edge of the 
main forest-zone has exposed them to penetration by the more vuile forms from 
the surrounding savannahs. 
B. Open Forest in Angola (Mt. Moco to Catengue). 
The collections made in the more open type of forest are, unfortunately, not 
sufficiently large to permit of any very extensive generalizations. But it seems 
probable that this vegetational zone has very little in common with the true 
Rain Forest. Of the 12 species collected, 55 per cent, are either widespread species 
(Causus rJwmheatus, Rana oxyrhynchus and Bufo regularis) or are savannah 
species found in the countries bordering the forest. The only two which might 
be regarded as Rain Forest forms are Chamaeleo dilepis subsp. and Ichnotropis 
biviitata (?) (q.v.), and it may be significant that both of these differ from the 
normal. Their presence, however, does suggest that this open forest may be a 
derivative of the primeval Rain Forest, which has been subjected to penetration 
from the savannahs, and in which conditions have changed to such an extent that 
the components of its original fauna have either been exterminated or become 
modified to meet the changing environmental conditions. 
C. Open Forest in Damaraland (Sissekab). 
Only 9 species were collected in this zone, but these suffice to indicate that 
it has little in common with the Angolan forests. Two of the species occur also 
in the damp forests of Angola, but both are tolerant forms, Bufo regularis and 
OerrJwsaurus flavigularis, whilst the only one which also occurs in the open 
forests of Mt. Moco is probably racially distinct {C'immaeleo dilepis, q.v.). The 
remainder are either species indigenous m the Damaraland region, or forms which 
inhabit the zone bordering the equatorial rain forests on the south and east. 
D. Dry, Granitic or Sandy Localities in Angola (Bocoio and Lobito) or Limestone 
(Morro de Pundo). 
Three species only were collected, the cosmopolitan Hemidactylus mabouia 
and two species indigenous in the dry country of S.W. Africa and southern 
Angola, Mabuia acutilabris and Rhoptropus boultoni. 
E. Dry Granite, Gneiss, Limestone or Sand in S.W. Africa (Lake Otjikoto, 
Waterberg Mts., Omongongua, Okahandja, Swakopmund, Windhoek, 
Hoffnung, Rehoboth, Naukluft Mts., Maltahohe, Voigtsgrund and 
Satansplatz). 
NOVITATES ZOOLOOICAE XL. 1!*IJ6. U7 
Of the 46 species collected in these localities, only 4 are found in the 
damp Angolan forests and these have been discussed above. The remainder 
consists of one or two tolerant species which range over almost the whole of the 
continent, irrespective of vegetational or climatic conditions, of a more numerous 
class of species which extend over a greater or less extent of the belt surrounding 
the Rain Forest and of species indigenous in the dry region from southern Angola 
to Little Namaqualand. These latter constitute the largest element of the fauna 
(46 per cent.), but it is quite impossible to distinguish clearly between them and 
those species which range across the southern border of the Rain Forest, and 
these, in turn, grade insensibly into others with a more extended range around the 
Rain Forest. Thus 21 species may be classed as strictly indigenous in the dry zone 
and are not known to extend eastwards beyond the Kalahari ; but another group 
of 6 species extends eastwards into the Transvaal and S. Rhodesia ; 7 others range 
still farther eastwards into Mozambique and Southern Tanganyika Territory ; 
8 have a similar range, but extend northwards into the Kenya, Sudan or Somaliland 
areas ; and only one, Kassina senegalensis, ranges completely round the whole Rain 
Forest. This offers a decided contrast with the conditions obtaining in the Angolan 
forests, where, of the 7 invading species, 5 range completely round the forest belt 
and the other two have an extensive range on both the south and east of this area. 
If, as seems probable, a process of desiccation is in progress in the south-west 
of Africa similar to that which is occurring in the north and north-east, the natural 
sequence of changes in the herpetological fauna consequent on the destruction of 
the primeval forest and its replacement by arid steppe and desert conditions may 
well be exemplified by the fauna of the different types of country considered above. 
First of all, with the approach of the dry zone to the original dense, wet, forest 
there is an infiltration of new species which are widely distributed round the margin 
of the dying forests (A). As the true Rain Forest is replaced by a more open type 
of savannah forest (B and C) the original species are, for the most part, extermin- 
ated and replaced by other species with a wide distribution in similar zones. As 
desiccation proceeds, greater and greater specialization becomes necessary to the 
fauna, and new forms make their appearance, which, since the area of maximum 
desiccation in the south-west is still relatively small, have a more and more 
restricted range. The intermediate stages of open forest have a fauna which, to 
judge from the present collections, is much more limited in the number of its 
component species. This may, of course, be a purely fictitious conclusion and 
he merely the expression of the length of time spent by the collector in each zone. 
But it may have a deeper significance, for Sanderson (I9.'56, p. 178) has recently 
shown that, in the Cameroons, the artificial clearing of primeval forest produces 
a similar result, and that in the various intermediate stages before the land is 
allowed to revert to its natural, j)ermanent, secondary conditiim, the numljcr of 
species in the frog fauna is very much smaller than in either the original or final 
stages. A possible explanation of this phenomenon may lie in the fixity, or other- 
wise, of the different zones. Both the primeval forest and the final condition of 
steppe and de.'iert after its desiccation may not be large and their actunl sizes are 
changing ; but their conditions are relatively permanent and they have fixed 
geographical centres. The intermediate zones, on the other hand, may be large 
in extent, though strip-like, but they are fluctuating and have no permanent 
positions. fV)iiso(juontly species which can siu'vive in them nuist not only be 
a<laptc(l til llic plw'sicnl mikI liiolngical eiiviinnnioiil , but must be possessed of 
118 
NOVITATES ZOOLOGICAE XL. 1936. 
the requisite mobility to maintain their positions in this shifting medium. Too 
great a dependence on external factors and too close a linkage with the environ- 
ment must tend to render a species immobile, for it can only persist when its 
exact requirements are fulfilled ; the more complex these are, the less is the 
probability of exactly the same conditions arising again within migrational range 
when the original combination is destroyed. But in areas of fixed position and 
with relatively permanent physical conditions, nicety of adjustment to the 
environment is no handicap. 
The following lists, in which the species are grouped roughly according to 
their geographical distribution, shows also their distribution in the different 
climatic and vegetational zones, as determined by the present expedition. It 
shows the changes mentioned above and also emphasizes the great differences in 
the composition of the forest and desert faunae. In the former snakes abound, 
whilst geckos and lacertid lizards are almost non-existent, whereas, in the dry 
zones, the number of snakes is greatly reduced, and the dominant elements in 
the lacertilian fauna are geckos and lacertids, with, also, an increase in the 
number of skinks of the genus Mabuya. In the Somaliland Peninsula, another 
recently desiccated area (Parker, 1932), the fauna is essentially similar in its 
general composition, with numerous geckos (the genus Hemidactylus taking the 
place of the S.W. African Pachydactylus), lacertids and skuiks. 
I. Widespread Species very tolerant of Different Climatic and Vegetationai. 
Conditions. 
1 A species with many local races. Aa noted below, the form in zone B is very different from 
that in zones C and E. 
IT. Species confined to the Rain Forest and its Outliers. 
Typhlops punctatus intermedins 
Lycophidion ornatiun s^p. n. 
Oophilosiium parheri 
Horrnonotus modestus. 
Chlorophis heterodermus 
Boiga pulverulenta 
Boiga hlandingi 
Thelotornis kirtlandii . 
Naja goldii 
Naja rnelanoleuca 
Dendraspis jamesoni . 
Bids yuisicornis . 
H€7t} idncUfl'.is loy^gicephnlus . 
Ichnotropis bivittata . 
Mahiif/a niacxdilahris 
Lygosown brevicep/t 
Lygosotna dewittei 
Ablepharus cabindae . 
Feylinia currori 
Chaynafileo etiennei 
Arthroleplis parvulus 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
NOVITATES ZOOLOGICAE XL. 1936. 
III. Species Indigenous in the Angolan Forested Region. 
119 
' Distribution uncertain ; Angolan specimens differ appreciably from those from Damaraland 
(see p. 132). 
IV. Species DisTHiBnrED RotrND the Periphery of the Whole Rain Forest, but 
encroaching upon it. 
V. Species bordering the Rain Forest on the East and South (from the Sudan- 
Somaltland-Kenya Region southwards). 
VI. Species borderino thk Rain Forest on the South, fhom Tanoanyika Territory 
OR Mozambique westwards. 
Lept/)typl)lop8 difttanti 
Aspiffrfnp.'i scutatua . 
Pfirh'/tifirfi/lus hihrotti turntri 
f'nrfn/'l'irfi/fu.t punrfatus 
Acontins melengris 
A mphiahnryia qtiadrifrons . 
Xenopus trtrvis Incria 
120 
NOVITATES ZOOLOGICAE XL. 1936. 
VII. Species confined to Countries to the South of the Forest, but not extending 
INTO Tanganyika Territory or Mozambique. 
Psammophis notostictus 
Psammophis bocagei . 
Bitis caudalia . 
Ptenopus garrulus 
Agania hispida aculeata 
Eremias lineo-oceWita 
Eremias lugubris 
Rana fuscigula Juscigula 
+ 
+ 
+ 
+ 
-r 
+ 
+ 
+ 
VIII. Species confined to the Dry Aeeas in Angola, 
AND NaMAQUALAND. 
Damakaland, Bechuanaxand 
Typhlops boylei . 
Rhoptropus barnardi 
Rhoplropus bovltoni . 
Packydactylns bibroni piditzerae 
Pachi/dactylus laevigatus 
Pachf/flaclylus weberi 
Padtydactylus rugosiis 
Pachydactylus purcelli 
Narudasia f estiva 
Choiidrodactylus anguUJer 
Agawa anchiefae anchietae 
Cordylosaurus trivittalus 
Zonurus jordani sp. n. 
Eremias namaqKensis 
Eremias undata 
Scapteira reticulata 
Nlicras intertexta damarana subsp. n. 
Mabuya sulcata 
Mabuya damarana 
Mabuya acutilabris . 
Mabuya binotata 
Monopeltis cupensis . 
Phrynom^rus annectens 
Bufo jordani sp. n. 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
REPTHilA SERPENTES. 
1. Typhlops punctatus intermedius, Jan. 1861. 
3 Congulu April 
1 Quirimbo May 
These specimens are of the lineate form \\ith the central zone of the belly 
immaculate ; the pigmentation does, however, invade the sides of the belly more 
than in normal intermedius, so that in this respect they approach typical pwwctofits. 
2. Typhlops schlegeh mucroso (Peters, 1854). 
1 Omongongua Jan. 
Scales 34 ; ratio of diameter to length, 27. 
3. Typhlops boylei Fitzsimons, 1032. 
1 Hoffnung Dec. 
This example differs from the original description of the species as follows : 
The rostral extends backwards to the level of tlie eyes, the nasal suture proceeds 
NOVITATES ZooLoaiCAE XL. 1936. J21 
from the first upper labial, instead of from the suture between the first and 
second, the .portion of the rostral visible from below is once and a half, instead of 
" slightly," broader than long, and there are 28 instead of 26 mid-body scale-rows. 
But in a series of 5 specimens from Ghanzi, near the type locality of hoylei, 
there is some variation in all these characters. The rostral may not extend quite 
to the level of the eyes ; the length of the first upper labial is not absolutely 
constant, so that the nasal suture may or may not touch the laljial suture, 
and there are from 26 to 28 scale-rows. The number of scale-rows and high 
diameter/length ratio suggest close afiinity with T. lalandii, and it may be that 
some S.W. African records of this may be in reality based on boylei ; the two 
are readily distinguishable by the much more prominent snout and cutting 
rostral edge of the latter. 
4. Leptotyphlops distant! (Boul., 1892). 
Slenostoma scniifrons {non Peters. 1854) Peters, IS65. Hon. Ak.. Berlin, p. 261. fig. 5 ; idem. 1882, 
Reise Mossarabique, iii, p. 104 (part) pi. xv, fig. 4. 
Olauconia scutifrons Boulenger, 1890, Ann. Mag. N.H. (6), vi, p. 92 ; idem, 1893, Cat. Snakes Brit. 
Mils., i, p. 68. 
Qlaucoitm hoeitqeri Werner, 1899. Zool. Am., xxii, 581. p. 116. 
Olauconia latifrons Sternfeld, 1908, Sitzber. Ver. Nat. Fr., Berlin, p. 94. 
Olauconia okahandjana Ahl, 1924, Archiv Naturg., xc, 4, 5, p. 247. 
1 Hoffnung Nov. 
3 „ Dec. 
2 „ Feb. 
1 Windhoek Jan. 
2 ? 
1 Okahandja Dec. 
Fitzsimons and Loveridge (Loveridge, 1933, p. 225) have given reasons for 
considering diManti Boulenger to be a synonym of scutifrons Peters. But in 
doing so thej' have not given any further consideration to the problem of whether 
scutifrons Peters 1854, is conspecific with the scutifrons of Peters 1865, and of 
Boulenger 1893 (p. 68). The specimen described and figured by Peters in 1865 
was that recorded by Boulenger, but this point seems to have been missed by 
Sternfeld and Werner, who have both discussed the matter. Sternfeld (1908, 
p. 94) considered that the true scutifrons of Peters (1854) was distinct from the 
scutifrons of Boulenger (which is also the scutifrons of Peters, 1865 nee 1854), the 
former lacking the anterior upper labial ; he accordingly propo.sed a new name, 
latifron.s. for the latter. Werner (1909, ]>. 21(1) is unconvinced that the two rcnlly 
are distinct, preferring to believe that a labial so small as that of latifrons might 
easily be lost (presumably as an individual anomaly), giving rise to the condition 
found in the type of scutifrons. Sternfeld (191(i, p. 13) maintains his original 
views without further Cduunents. and Werner (191(1, p. 354), with UKue material 
before him, is still dubious, but speaks of a " scutifronx group " with 4 members, 
distinguished thus ; 
First Supralabial Supraoculars 
Present Present latifrons Sternfeld. 
Present Absent horttgeri Werner. 
Absent Present sciitifron.f Pet^'rs. 
Absent Absent labialis StenifcM. 
122 ' NOVITATES ZOOLOGICAE XL. 1936. 
There can be little doubt that Loveridge and Fitzsimons are correct in con- 
sidering scutifrons, sensu Peters 1S65, and Boulenger 1893, i.e. latifrons Sternfeld, 
to be synon3Tnous with distanti Boulenger. But the position of true scutifrons 
is by no means established, and it may, perhaps, be significant that all the examples 
of scutifrons and labialis which have been recorded are symmetrical ; if their 
labial condition was merely anomalous, some asymmetry might have been 
expected. Accordingly, until some proof is forthcoming that the absence of the 
anterior labial is an individual abnormality, L. scutifrons must be considered as 
a species distinct from L. latifrons ; the latter is conspecific with distanti, which 
is the older name and must be be used. L. labialis and L. boettgeri, both known 
from one or two specimens only, are probably based on individual aberrations of 
scutifrons and distanti respectively, whilst yet another name, G. okahandjana 
Ahl (1924), ought, probably, to be added to the synonymy of the latter. The 
topotype in the present collection agrees well with the original description except 
that the ratio of length to diameter Ls 69 instead of about 53 ; but Werner (1910, 
p. 354) records a variation in " scutifrons " of from 55 to 105. 
5. Boaedon lineatus Dum. & Bibr., 1854. 
$ Okahandja Dec. -Feb. 
2 (^(J, $, 2 juvs. Congulu April 
6. Lycophidion omatum sp. n. 
Among the collections from Congulu are two specimens of a species of 
Lycophidion closely allied to L. capense. They differ constantly from a large 
series of the latter, including 5 others from Angola, in having a broad light band 
bordering the snout (as in L. capense uzungwense) , in a frontal as broad as, or 
broader than long, and in having the posterior nasal separated from the first 
upper labial. Each one of these differences appears trivial in itself, but they are 
correlated, for exactly the same differences were found in four other examples 
from Uganda, and in 10 specimens from the Belgian Congo. This suggests that 
they represent a distinct species, almost intermediate between L. capense and 
L. laterale. 
The holotype is a female in the British Museum, from Congulu, Angola ; 
collected in April 1934 by Dr. Karl Jordan. 
Diameter of the eye greater than its distance from the lip. Rostral more 
than twice as broad as deep ; internasals about as large as the nasals ; prefrontals 
longer than broad ; frontal little broader than long, a little shorter than its 
distance from the rostral, | the length of the parietals ; loreal twice as long as 
deep ; one preocular, as large as the supraocular and making a broad suture 
with the frontal ; two postoculars, both in contact with the parietal ; temporals 
1 + 2 ; eight upper labials, the first separated from the posterior nasal, and the 
third, fourth and fifth entering the eye. Two pairs of small chin-shields, the 
posterior the smaller ; five labials in contact with the anterior. Scales smooth, 
with single apical pits in 17-17-17 rows ; ventrals 199 ; anal entire ; subcaudals 
42 -f 1. 
Grey-brown above, each scale faintly mottled with lighter ; a broad, light 
band I'ound the snout, extending backwards on to the temple, where it becomes 
indistinct ; lower surfaces grey, the chin and the posterior edge of each scute 
lighter. 
NOVITATES ZoOLOOITAE XL. 1936. 123 
Length from snout to vent 299 mm. ; tail 45 mm. 
The paratypes are : 
B.M. 
Congulu. ?. Sc. 17-17-17. V. 199. C. 46 + 1. 
Mus. Congo 5174 
Nyonga, Katanga. (J. Sc. 17-17-17. V. 196. C. 48 + 1. 
Mus. Congo 4952 
Nyonga, Katanga, juv. ^J. Sc. 17-17-17. V. 205. C. 43 + 1. 
Mus. Congo 1925 
Karemi, L. Tanganyika. ?. Sc. 17-17-17. V. 194. C. 40 + 1. 
Mus. Congo 4000 
Usumbura, L. Tanganyika. ?. Sc. 17-17-17. V. 200. C. 41 + 1. 
Mus. Congo 3823 
Ki.ssenyi, Kivu. juv. ^. Sc. 17-17-17. V. 187. C. 44 + 1. 
Mus. Congo 3797 
Lulenga, Kivu. juv. (J. fSc. 17-17-17. V. 174. C. 46 + 1. 
Mus. Congo 3793 
Lulenga, Kivu. ?. Sc. 17-17-17. V. 190. C. 39 + 1. 
Mus. Congo 3781 
Lulenga, Kivu. $. Sc. 17-17-17. V. 194. C. 39 + 1. 
Mus. Congo 1144 
Beni, Ituri. 9. Sc. 17-17-17. V. 198. C. 41 + 1. 
Mus. Congo 1688 
Moera, Ituri. <^. Sc. 17-17-17. V. 196. C. 53 + 1. 
B.M. 1934.12.15.555-556 
Muko, 7,000 ft., Kigezi, Uganda. $?. Sc. 17-17-17. V. 202, 204. 
C. 38 + 1, 37 + 1. 
B.M. 1934.12.15.557 
Kayonsa Forest, 7,000 ft., Kigezi. ?. Sc. 17-17-17. V. 200. C. 39 + 1. 
B.M. 98.12.27.17 
Mau Ravine, 7,500 ft., Uganda. ^. Sc. 17-17-17. V. 198. C. 46 + 1. 
Total variation in pholidosis : Sc. 17-17-17. V. ^^ 174-205. $? 190-204. 
C. cJ(? 43-53 + 1. ?? 37-42 + 1. 
The series shows singularly little variation in colour or in the proportions of 
the head-shields. With respect to the latter and to the number of ventrals and 
subcaudals it is very similar to L. lalcrale, but may be at once distinguislied by 
the single ajiical pits. 
7. Oophilosituni parkeri Angel, 1934. 
2 <;JcJ, 4 $? Congulu April 
These six specimens are referred to this species with an element of doubt. 
The species was originally described as differing from O. fa-'^rinhitii in lia\ing 
fewer teeth, longer parietals and only six upper labials, of which two entercfl the 
eye. AnoUier difference was found in the size of the maxillary foramen, but this, 
quite correctly, was considered of doubtful importance. Examination of a much 
larger series of O. fnxn'nlvm than was available when the geiuis Onphilosiliim was 
described (I'lirker. 1!t33, j). 545) reveals that the niaxilhirv for.inieii !.•< of no 
importance whatever from a taxonomic st;vii(l|iiiinf , thiil the l(-n;;11i of the 
124 
NOVITATES ZOOLOGICAE XL. 1936. 
parietals of O. fasciatum varies sufficiently to include the condition of O. parkeri 
and that there is a greater variation in the number of teeth than was previously 
believed. The size of the maxUlary foramen is, to some extent, correlated with 
age ; no very young specimens have been found in which it is not very large, but, 
on the other hand, it has been found to exist in specimens considerably larger 
than others which lack it. Possibly if large series from different localities were 
available its absence might prove to be an age character, but the age at which it 
closes varies in different areas. The number of posterior maxillary and mandi- 
bular teeth is extremely variable, and it is rather remarkable that the number in 
one jaw appears to be quite unrelated to that in the other. The only remaining 
characters whereby the species parkeri might be recognized are the lower number 
of upper labials and the fact that only two, instead of three, enter the eye. The 
present series is quite uniform in this respect and agrees with another example 
from the Ituri, which is in the same general area as the type locality of parkeri. 
If these really are conspecific the variation in the development of the maxillary 
foramen and the number of teeth is quite comparable to that found in 0. fascia- 
tum. The variation in these characters in the two species is : 
B. O. parkeri. 
Scales about the body are 17-17-15 or 17-17-17 ; 
174, (JcJ 170, ISO ; Subcaudals ?? 40, 30, 39 and 37 
Ventrals ??171, 172, 173, 
+ 1, c?c?43, 43 +1. 
NOVITATES ZOOLOGICAE XL. 1936. 125 
8. Honnonotus modestus (Dum. & Bibr., 1854). 
$ Congulu April 
This example appears to be the first recorded from Angola, but is quite 
typical of this widespread Rain Forest species. 
9. Chlorophis heterodermus Hallowell, 1857. 
2 $$, 5 juvs. Congulu April 
10. Chlorophis irregularis (Leach, 1819). 
$ Mt. Moco March 
2 juvs. Congulu April 
Schmidt (1923, p. 76) has already drawn attention to the fact that this species 
of the Savannahs does occasionally invade the forest area in the Cameroons and 
the Ituri district ; its occurrence in the forested Congulu area and on Mt. Moco 
indicates a similar encroachment in the south. 
11. Dasypeltis scaber (Linn, 1758). 
3 juvs. Congulu April 
12. Tarbophis semiannulatus (Smith, 1849). 
juv. Okahandja Feb. 
13. Boiga pulverulenta (Fischer, 1856). 
cj Congulu April 
14. Boiga blandingi (Hallowell, 1844). 
cJ, $ Quii-imbo May 
15. Crotaphopeltis hotamboeia (Laurenti, 1768). 
(J, juv. Quirimbo May 
16. Thelotomis kirtlandii (Hallowell, 1844). 
2 $? Quirimbo May 
Both examples lack the head-markings so frequently found in southern and 
eastern specimens and thus conform to tlic typical forest race. 
17. Psammophis notostictus Peters, 1867. 
o Maltahohc Dec. 
$ Hoffnung Jan. 
The male is aberrant in having U upper labials, of which the fourth, fifth and 
sixth enter the orbit. 
126 NOVITATES ZOOLOGICAE XL. 1936. 
18. Psammophis bocagii BouL, 1895. 
$ Otjosongombe Nov. 
This appears to be the most southerly record for the species, which ranges 
into Angola and eastwards through Bechuanaland to S. Rhodesia. 
19. Naja goldii Boul., 1895. 
2 $$ Quirimbo May 
The discovery of this Rain Forest species in Angola extends its known range 
considerably ; it has not previously been recorded south of the lower Kasai 
River and Lower Congo. Both specimens are, however, quite typical with 15 
scale-rows, 194-198 ventrals and 78-80 + 1 subcaudals ; the larger measures 
1,770 mm. 
20. Naja melanoleuca Hallowell, 1857. 
$ Quirimbo May 
juv. Congulu April 
21. Dendraspis angusticeps (Smith, 1849). 
Head Sissekab Nov. 
22. Dendraspis jamesoni (Traill, 1843). 
$ Quuimbo May 
juv. Congulu April 
23. Aspidelaps scutatus (Smith, 1849). 
?, juv. Okahandja Feb. 
$ Omongongua Jan. 
Fitzsimons (1935, p. 326) has drawn attention to the uniformly dark heads 
of western (Kalahari and S.W. Africa) examples of this species, as contrasted 
with the white-blotched heads of eastern specimens. The two adults in the 
present collection agree with this generalization, but the juvenile has white 
blotches disposed exactly as in a cotype from Natal. 
24. Bitis caudalis (Smith, 1849). 
(J, ?, juv. Hoffnung Jan. 
(J Biillsport Dec. 
juv. Sissekab Nov. 
25. Bitis nasicomis (Shaw, 1802). 
(J, 2 $9 Quirimbo May 
This constitutes yet another first Angolan record of a typical species of the 
Rain Forest. Previously it has not been reported south of the Lower Congo. 
26. Causus rhombeatus (Licht., 1823). 
2 $? Mt. Moco March 
juv. Quirimbo May 
NOVITATES ZOOLOOICAE XL. 1036. 127 
27. Causus resimus (Peters, 1862). 
2 (J (J, juv. Quirimbo May 
juv. Congulu April 
The distribution of this species is rather puzzling. It is certainly not a forest 
species, but is common in the eastern savannahs from the Anglo-Egyptian Sudan, 
Ethiopia and Somaliland southwards to Tanganyika Territory. But it does 
not appear to have been recorded between the latter and Angola. Boulenger's 
record of the species from Rhodesia (1907, p. 12), quoted by Pitman (1934, 
p. 300), is erroneous and based on examples of C. defilippii. In addition to this 
apparent discontinuity of range is the fact that in Angola the snake appears to 
be confined to the low-lying, swampy and forested littoral zone (Bocage, 1895, 
p. 146). 
SAURIA. 
28. Rhoptropus bamardi Hewitt, 1926. 
$ Sissekab Nov. 
3 (J (J, 4 ?? Lake Otjikoto Nov. 
This series agrees with other examples in the British Museum from the 
Messum River and from Mossamedes ; they have the acute snout so characteristic 
of bamardi, none have any transversely enlarged plates beneath the tail and the 
total number of lamellae beneath the fourth toe varies from 15 to 17. 
29. ? Rhoptropus boultoni Schmidt, 1933. 
(J Bocoio, Benguela March 
This single specimen differs from the preceding series in having a broader, 
more rounded snout, a series of transverse plates beneath the tail on all except 
the first three segments, rather more subdigital lamellae (20) and smaller chin- 
shields. It agrees well with the description oi bradfieldi Hewitt, except that the 
scales on the snout are faintly keeled in the canthal region, and in having the 
second pair of lower labials distinctly elongate. But, at the same time, it is 
obviously conspecific with 5 other specimens from Benguela, which are now in 
the British Museum, and these show more deviations from the description of 
bradfieldi in having chin-shields, 4 to 6 of the anterior caudal segments without 
transverse plates and fewer (17-20) lamellae beneath the fourth toe. If these 
northern specimens really are conspecific with bradfieldi, the species exhibits a 
range of variation which would include the described condition of boultoni 
Schmidt, and if these two are synonymous then the described species of the 
genus (excluding bracconnieri of uncertain status) may be distinguished thus : 
I. Median gular scales larger than those on the belly ; anterior nasals separated 
by two or three granules ; digits very long and slender ; no preanal pores. 
R. afer Peters. 
(C and S. Damaraland.) 
II. Median gulars much smaller than the ventral scales ; anterior nasals separated 
by a single scale ; tligits shorter and stouter ; preanal pores usually 
present. 
128 NOVITATES ZOOLOGICAE XL. 1936. 
A. Tail without transversely enlarged plates below ; snout pointed ; 
siibdigital lamellae beneath the fourth toe 14-17. 
R. barnardi Hewitt. 
(N. Damaraland to Mossamedes.) 
B. Tail with transversely enlarged plates below, at least posteriorly ; 
snout rounded ; subdigital lamellae beneath the fourth toe 17-23. 
B. boultoni Schmidt. 
(N. Damaraland and Benguela.) 
30. Hemidactylus longicephalus Bocage, 1873. 
7 (Jc?, 9 ?? Congulu April 
3, 2 $$ Quirimbo May 
31. Hemidactylus mabouia (Mor. de Jonnes, 1818). 
(J Lobito April 
Pachydactylus bibroni (Smith, 1849). 
The various species and races forming the bibroni complex are at present 
very little understood. Boulenger (1910), in his survey of the South African 
forms, recognized two species, bibroni and laevigatus ; stellahm Werner was not 
considered and Gray's turneri had long been considered a synonym of bibroni. 
Werner, in the same year, recognized bibroni, laevigattis, stellaius and boulengeri 
(of Tanganyika Territory) as distmct species. Hewitt considered both laevigatus 
and stellatus to be merely subspecies of bibroni, giving the ranges of the three 
respectively as S.W. Africa, Great Nama(qua)land and the Cape Province. But 
Schmidt (1933), finding laevigatus in the same localities as specimens of bibroni, 
confessed his inability to understand the distribution of the two, and of 
stellatus, on the hypothesis of their all being subspecies, and so accorded them 
all full specific rank, but described the Angolan bibroni as a new subspecies, 
p^diizerae. 
It is evident, from a survey of the material in the British Museum, that 
one probable cause of confusion is misidentification owing to the fact that 
the degree of stellation of the dorsal tubercles, usually regarded as diagnostic 
of stellatus, is largely an age-character, and also occurs in typical bibroni and, 
to some extent, in laevigatus. A tentative arrangement, which seems to 
overcome the distributional difficulties, is to regard stellatus and pulitzerae 
as races of bibroni, and laevigatus as a distinct species. Unfortunately the 
name " stellatus " has to give way to the much older turneri, since the 
subspecies with the stellate tubercle is found to range from Damaraland to 
Mozambique. The various forms are not readily separable, but the following 
key may be of assistance : 
I. Dorsal tubercles on the middle of the back smooth, or very obtusely keeled ; 
stellate tubercles confined to the region behind the ear. Nostrils directed 
almost vertically upwards. Gular scales flat, half the size of the ventrals. 
Damaraland. P. laevigatus Fischer. 
NOVITATES ZOOLOQICAE XL. 1936. 120 
II. Dorsal tubercles always strongly keeled and trihedral. Nostrils lateral. 
Gular scales almost granular. 
A. Stellate tubercles confined to the back of the head and flanks ; dorsal 
tubercles sometimes with additional radiating keels on their 
posterior facets. Distance from snout to anterior border of orbit 
no longer than the distance from the eye to the posterior border 
of the ear in the adult. Cape Province and Namaqualand. 
P. bibroni bibroni (Smith). 
B. Mid-dorsal zone with stellate tubercles, the radiating keels being 
present on the lateral as well as the posterior facets of many of 
them. Snout as in A. 
Damaraland, Orange Free State, Transvaal, Rhodesia, Portu- 
guese East Africa, Nyasaland and Tanganyika Territory. 
P. bibroni turneri (Gray). 
C. Tubercles as in A. Distance from the tip of the snout to the anterior 
border of the orbit much longer than the distance from eye to ear. 
Angola. P. bibroni pulitzerae Schmidt. 
32. Pachydactylus bibroni tumeri (Gray, 1864). 
. Soc. London, p. 59, pi. ix, fig. 2. 
.910, Jena Denkschrift, xvi, p. 309. 
Otjosongombe, 1,600 m. Nov. 
Homodactylm tumeri Gray, 1864, Proc. Zool. Soc. London, p. 59, pi. ix, fig. 2. 
Pachydactylus Hbroni var. stellatiis Werner, 1910, Jena Denkschrift, xvi, p. 309. 
This series has been compared with the cotypes of turneri and cannot be 
distinguished by any characters which might be of specific importance. Both 
have the stellate tubercles which characterize the rather ill-distinguished northern 
race of P. bibroni, but there is a considerable amount of individual variation in 
this respect. In juveniles the stellate tubercles are confined to the region behind 
the ear and the posterior part of the flanks, in which areas they are present in the 
typical form ; but with increasing age they are developed more and more over 
the middle of the back. Specimens have been examined from Tanganyika 
Territory, Nyasaland, Portuguese East Africa, Rhodesia (North and South) and 
the Transvaal. Other examples from the Orange Free State and Bechuanaland 
cannot be referred with confidence either to turneri or the typical form. 
33. Pachydactylus bibroni pulitzerae Schmidt, lit3.'5. 
3 (5(5, 1 $,4 juvs. Morro de Pundo, Angola May 
These 8 specimens, together with 4 others from the province of Bengucla, 
resemble typical bibroni (cotypes examined) in the degree of development of their 
dorsal tubercles, but differ from that form and from turneri in a longer, more 
pointed snout, a somewhat narrower interorbital space and more pronounced 
frontal concavity. These differences seem to indicate the existence of a distinct 
Angolan race for which the name pulitzerae is available. 
<J 
130 NOVITATES ZOOLOGICAE XL. 1936. 
34. Pachydactylus laevigatas Fischer, 1888. 
,(J Windhoek Jan. 
(^, 2 juvs. HofFnung, near Windhoek Feb. 
2 cJcJ, 2 ??, 2 juvs. „ „ Dec. 
<S .. .. Jan. 
cJ, $ BiiUsport, Naukluft Mts., 1,450 m. Dec. 
2 juvs. Rehoboth, 1,450 m. Dec. 
3 cJ(J, 2 juvs. Satansplatz, 1,300 m. Dec. 
35. ? Pachydactylus weberi Roux, 1907. 
(J Windhoek Jan. 
(J Hoffnung Dec. 
Recently Hewitt (1932, p. 124, and 1935, p. 315) has cast doubts on the 
validity of this species and on the accuracy of nearly all records of the species 
except the original one. He considers true iceberi (from KlipfonteLn, Namaqua- 
land) to be a subspecies of P. capensis ; specimens from Garies and nearby locali- 
ties in the south of Little Namaqualand are referred to another subspecies, 
P. capensis gariesensis, whilst specimens from the Khan River, Karibib and Keet- 
manshoop (i.e. from Damaraland and Great Namaqualand) are determined as yet 
a third subspecies, P. c. werneri. The present specimens must almost certainly 
belong to the same race as those from Karibib and also those from Windhoek 
recorded as iveberi by Sternfeld (1911, p. 387, and 1911a, p. 14), yet they do not 
agree in detail with the descriptions of any of the so-called races. This strongly 
suggests that some of the latter will prove to be untenable and geographical 
considerations suggest that werneri = weberi ; since proof of intergradation with 
capensis is still lacking, it is preferred to use the name specifically. 
36. Pachydactylus rugosus Smith, 1849. 
J Naukluft Mts., 1,300 m. Dec. 
This specimen has the colour pattern and mental shield ascribed to P. rugo- 
sus f rater Hewitt. But none of the other characters used to define that race 
are at all apparent. The gular scales are conical and the ventrals pyramidal, as 
in the type. 
37. Pachydactylus purcelli BouL, 1910. 
2 cJcJ, 2 ?? Maltahohe, 1,460 m. Dec. 
38. Pachydactylus punctatas Peters. 
(J, § Hoffnung, near Windhoek, 1,850 m. Feb. 
It seems very probable that none of the so-called subspecies of punctatus 
are recognizable. Fitzsimons (1935, p. 339) has already expressed the ojjinion 
that brunnthaleri Werner, described from S. Rhodesia and also reported from 
Angola (Schmidt, 1933, p. 5), is untenable. The present two examples only differ 
from others from Rhodesia and Angola in having the first labial entering, or very 
narrowly separated from, the nostril, in which character they approach langi 
Fitzsimons and amoenoides Hewitt. 
NoVITATES ZOOLOOICAE XL. 193(5. 1 ;^ j 
Althougli the two were collected on the same farm they are very different in 
colour, suggesting that colour differences, such as are said to characterize langi 
and bicolor Hewitt, are of little consequence. 
39. Narudasia festiva Meth. & Hewitt, 1913. 
$ Sataiisplatz, 1,300 m. Dec. 
This example has been compared with a paratype of the species and found 
to agree well. The status of the genus, however, is open to question. The 
original describers compared it with Homonota and Stenodaclylus, but in reality 
it appears to be almost indistinguishal)le from Gymnodaclylus. It differs from 
the majority of the species of that genus in the absence of chin-shields, but that 
is by no means a constant character and is insufficient to warrant the retention 
of a separate genus. But Gymnodaclylus is a cumbersome, and possibly not a 
monophyletic, assemblage which may ultimately be broken up ; any revision of 
the genus should certainly consider the status of Narudasia. 
40. Ptenopus garrulus (Smith, 1849). 
(J Mariental to Kehoboth Dec. 
41. Chondrodactylus angulifer Peters, 1870. 
cJ Biillsport, Naukluft Mts., 1,450 m. Dec. 
42. Agama anchietae anchietae Bocage, 1896. 
(J Voigtsgrund Dec. 
2 (JcJ, ? Satansplatz Dec. 
2 cJcJ, 2 ?? Maltahohe Dec. 
3 cJcJ> 6 ?? Windhoek Jan. 
2 $$ Hoffnung, near Windhoek Dec. 
In this series, especiallj' amongst those collected at Windhoek, there is every 
gradation between examples with a distinct dorsal crest and specimens without a 
trace of it ; at the same time the number of enlarged scales on the dorsum varies 
enormously, and in a few instances they are almost completely absent. Leg- 
length also varies and there is thus every stage of intergradation between A. 
anchietae anchietae, typically found in Angola, and A. anchietae methueni of 
Namaqualand ; Boulenger and Power (1921, p. 209) record the typical form from 
Maltahohe (misspelt " Matahiile "). 
43. Agama hispida aculeata Merrem, 1820. 
132 
NOVITATES ZOOLOGICAI; XL. 1936. 
7 <?(?, 6 ?? 
4 juvs. 
8 cJ<J, 5 ?$, 13 juvs. 
c?, 5 9? 
44. Agama planiceps Peters, 1862. 
Windhoek Jan. 
Morro de Pundo May 
Congulu April 
Quirimbo April 
Comparison of the Windhoek series with those from Angola shows that the 
former has, on the average, larger scales. This is most noticeable on the upper 
arm and femur, but is most easily described in terms of the number round the 
body. The Damaraland specimens available to the author (15) have a minimum 
mid-body count of 63 and a maximum of 74, the average being 69-5. The types, 
from Neu Barmen, in Damaraland, had from 73 to 76. But in Angola the 
number is much higher. In 55 specimens collected between 13° S. and 8° 30' S., 
the minima and maxima respectively are 71 and 99, with an average of 87-5. 
There is thus an overlap, and tabulation of the figures suggests that further 
material from southern Angola and northern Damaraland will show a continuous 
gradation, rendering it impossible to recognize any clearly defined races. The 
following localities are arranged in order from south to north. 
45. Agama atricollis Smith, 1849. 
$ Mt. Moco, 15-1,900 m. March 
Schmidt (1919, p. 477) has pointed out how this species, which is essentially 
a lizard of the savannahs of East and South Africa, enters the Rain Forest in the 
eastern Belgian Congo. Consequently its occurrence in the forest islands of 
Angola is to be expected. It should be pointed out that Schmidt's scale counts 
(loc. cit.) have been accidentally diminished by a hundred. In the present 
example the scales from chin to anus number 157. 
46. Cordylosauras trivittatus (Peters, 1862). 
5 (JcJ, 4 $$, 3 juvs. Hoffnung Dec.-Feb. 
2 ?$ Voigtsgrund, 1,300 m. Dec. 
Dr. Jordan reports that this lizard runs with lateral undulations, recalling a 
slow-worm or a snake, and emphasizes the very fragile nature of the tail. 
The series makes it very doubtful whether the subspecies aiislralis Hewitt, 
from Naraaqualand, can be maintained. This race was based on two examples 
only, with but a single other example for comparison, a series which was quite 
inadequate for the piu-pose. The present longer series, from the area in which 
the species was originally discovered, are obviously all conspecific, but show 
NOVITATES ZOOLOOICAE XL. 1936. J3;{ 
variations which cover almost every feature said to distinguish au-stralis and the 
other examples mentioned by Hewitt (1932, p. 115) which he considered might be 
specifically distinct. Thus, the light dorso-lateral lines never occupy more than 
two scale-rows on the body, though the light colour usually extends on to the 
posterior upper temporal. The parieto-occipital shield is completely undivided 
in 4 adults, has more or less distinct traces of incomplete sutures in 7 and is com- 
pletely divided into parietals, interparietal and frontoparietals in the 3 juveniles. 
Femoral pores vary from 7 to 9, of which the proximal 3-5 are well developed and 
the distals rudimentary in females, but all, except sometimes the end pore, are 
well developed in males. The tympanic shield is always broader than the 
posterior upper temporal. The only difference which is not covered by the 
present series is the keeling of the scales ; in the lumbar region the median keel 
is always the most prominent. The largest example, a female, measures only 
143 mm. from snout to vent. 
47. Gerrhosaurus flavigularis nigrolineatus -Hollowell, 1857. 
juv. Windhoek Jan. 
2 ads. Sissekab Nov. 
5 juvs. Congulu April 
6 ads. Quirimbo May 
The three examples from the dry country of Damaraland approach the 
typical form of East Africa in several respects, notably in their longer, narrower 
frontals. Those from the forested parts of Angola appear to be typical of tli^e 
West African subspecies. 
48. Zonurus jordani sp. n. 
Z. polyzonm {tum Smith) Boettger, 1894, Ber. Scnck: Xatf. Ges., p. S9 ; Werner, 1910, Jena Denkschrifl, 
xvi, p. 325. 
Holotype a $ from Hoffnung, near Windhoek, collected by Dr. K. Jordan 
in December 1933.^ 
Boettger, in recording a Zonurus from Rehoboth, under the name polyzonus, 
notes that it has only 7 femoral pores, and Werner records two others from 
Hereroland with pores varying from 5 to 7. The latter author calls particular 
attention to the fact that this low number is unusual, and that specimens of 
polyzonus from Namaqualand have from 12 to 17. Power, too, must have been 
suspicious of these records with a low number of pores, for in his revision of the 
genus (1930) he neglects them complotely. The single specimen collected by the 
present expedition apparently belongs to the same form, for it " keys " into 
polyzonus, but has the lower number of femoral pores, and certain other differ- 
ences, which seem to indicate that it is specifically distinct. The species does 
not appear to have received a name, and the type may be described thus : 
Frontonasal separated from the rostral by the supra-nasals which form a 
long suture ; lower eyelid with a transparent disc composed of two much-enlarged 
scales ; head scales rugose and disposed as in polyzonus, but the scal6s of the 
temporal region distinctly larger. Dorsal scales feebly keeled ; laterals similar 
in size but strongly keeled and miicronate ; 3<> transverse rows of scales between 
occiput and base of tail, a single row at the middle of the body containing 34. 
Ventrals smooth, in 22 longitudinal series, separated from the dorsals by 1 to 2 
rows of granules lying in a distinct fold. (Jular scales flat. Caudal scales very 
• I'arnlyiH's ii ^ mul n jiiv. from OtjoHunBcinilH', I'Vli. liCK! (\V. Hiicsili); phulidosis ami counts as 
in holutypf. 
134 NOVITATES ZOOLOGICAE XL. 1036. 
large and spinose, a single transverse series in eacli whorl. Scales on the limbs 
above strongly keeled and mucronate. 
Pale brown above, with small, obscure, darker spots. Lower surfaces 
uniform, pale, straw-colour. 
Length from snout to vent 111 mm. ; tail incomplete ; fore-limb 42 mm. ; 
hind-limb 56 mm. 
The species is closely alhed to polyzonits, with a large series of which (includ- 
ing the types) it has been compared, but also has some features in common with 
cordylus. From the latter it is readily distinguished by the presence of a supra- 
nasal and by its smaller scales, and from polyzonus as follows : 
jordani polyzomis 
(1) Femoral pores 5-7. (1) Femoral pores 12-17. 
(2) Temporals larger. (2) Temporals smaller. 
(3) Caudal whorls composed of a single (3) Caudal whorls at the middle of the 
row of scales. f tail composed of 2 rows of 
scales. 
(4) Posterior upper femoral scales (4) Posterior upper femorals scarcely 
much enlarged and strongly larger or more spinose than the 
spinose. anterior. 
(5) Transverse dorsal scales in 32-37 (5) Transverse dorsal scales in 38-46 
49. Eremias namaquensis Dum. & Bibr., 1839. 
J, 5 99 Hoffnung, 1,850 m. Nov.-Feb. 
2 cJcJ, 9 Rehoboth, 1,450 m. Dec. 
The females collected at Hoffnung in November, December and January are 
pregnant, but the single specimen taken in February has the oviducts still 
enlarged but empty, and the gonad itself is shrunken. 
50. Eremias undata (Smith, 1838). 
51. Eremias benguelensis Bocage, 1867. 
juv. Catengue, Angola March 
52. Eremias lineo-ocellata Dum. & Bibr., 1839. 
NoviTATES ZooLooicAE XL. 1936. i;j5 
53. Eremias lugubris (Smith, 1838). 
3 c?c?. ? HofFnung Jan. 
54. Scapteira reticulata Booage, 1867. 
2 c?c?> 2 juvs. Swakopmund (Sea Shore) Feb. 
55. Nucras intertexta damarana subsp. n. 
3 (JcJ, 5 ?? Sissekab Nov. 
There i.s still a great deal of confusion regarding the species and subspecies 
of N. intertexta and N. tessellata, but the above specimens appear to differ con- 
stantly from any of the races previously described, in their smaller size, the 
reduction of the occipital scale and the shortening of the interparietal, so that the 
parietals always form a suture behind it. In 35 out of 37 specimens of intertexta 
subspp. the interparietal separates the parietals completely and is truncate behind, 
where it forms a suture with the occipital or its rudiments. The Sissekab speci- 
mens probably represent a race with a very limited distribution, for Sissekab is 
one of the few localities in S.W. Africa where open forest country persists. This 
race may be described as follows, the description being drawn up from the 8 
cotypes. 
Head small, broader than deep (1-1-1-3), once and a half to once and two- 
thirds as long as broad ; its length contained 4 to 4-9 times in the length from 
snout to vent. Limbs moderate, the hind-limb reaching the wrist or the elbow ; 
foot as long as the head. Tail once and a half to twice as long as the head and 
body. 
Head-shields as in N. intertexta except that the frontal may be slightly 
shorter than its distance from the tip of the snout ; the parietals are only once and 
a half (vice If) as long as broad ; the occipital is rudimentary or, usually, quite 
absent ; the interparietal is shorter, forming an acute angle posteriorly (instead 
of being truncate) and the parietals form a suture behind the interparietal. A 
parietal foramen and pterygoid teeth are present ; 25 to 31 gular scales in a 
median series between chin-shields and collar ; latter composed of 8 or 9 scales 
of which the median is usually much the largest. Dorsal scales smooth, in 35 to 
41 rows at the middle of the body ; ventrals in 8 longitudinal and 28-34 trans- 
verse series. Femoral pores 10-13. Subdigital lamellae beneath tlie fourth 
toe 20-24. 
Dark brown aliove, with three narrow white stripes ; flanks with two narrow 
white lines of which the upper, commencing at the middle of the ear, is complete. 
Limbs with circular white spots above ; a white line along the back of the thighs 
and the inner l)orders of the tibiae. Uniform pinkish white beneath. 
The largest specimens, two females, measure 54 mm. from snout to vent ; 
another female of 52 mm. is gravid. The largest male is 52 mm. from snout to 
vent. 
56. Ichnotropis bivittata Bocage, 1866. 
I juv. Mt. Moco, Angola, 1,500-1,900 m. March 
This identification is questionable. The specimen has smaller scales than 
normal (45-56 at mid-body), but is too young to afford any reliable evidence as 
to whetlier racial differentiation has taken place outside the true Rain Forest. 
136 NonTATES ZOOLOGICAE XL. 1936. 
7. Mabuya bocagii Boul., 1887. 
4 Congulu April 
In all these four examples the scales about the body number 40 instead of 
36-38 and the dorsals are, for the most part, feebly tricarinate. Five keels do 
appear, however, in some places, and in other characters there is nothing whereby 
they may be distinguished from typical bocagii, so that the differences must be 
regarded as within the range of variation of the species. 
58. Mabuya striata (Peters, 1844). 
2 Otjosongombe Nov. 
2 Windhoek Jan. 
1 Hoffnung Jan. 
1 Voigtsgrund Dec. 
1 Mariental to Rehoboth Dec. 
2 Bullsport Dec. 
Fitzsimons (1935, p. 371) has recently drawn attention to the high propor- 
tion of specimens of this lizard which, in Bechuanaland, have the subocular 
reaching the edge of the lip. The same condition occurs in 20 out of the 21 speci- 
mens from Angola and Damaraland which have been examined. In the Trans- 
vaal and Rhodesia the proportion falls, somewhat, to 77 per cent. (49 examined), 
and in Mozambique and Natal there is a sharp decline to 6 per cent. (17 
examined). To the north, however, the decline is less rapid, specimens from 
Nyasaland, Tanganyika Territory and southern Kenya Colony showing about 
equal numbers of those with the subocular reaching and cut off from the lip. 
In Uganda, Ethiopia, Somaliland and the Sudan, however, the proportions are 
simUar to those in Mozambique and Natal, only 7 per cent. (32 examined) 
having the subocular bordering the mouth. 
59. Mabuya sulcata (Peters, 1867). 
M. sulcata var. sexstriata Werner, 1910, Jena Denkschrifl, xvi, p. 345. pi. viii. fig. 10. 
3 (J(J, 2 ?? Windhoek Jan. 
4 cJ(J, 4 $? Hoffnung Dec. 
1 juv. „ Feb. 
cJ Rehoboth Dec. 
2 cJ(J, $ Maltahohe Dec. 
cj, ? Voigtsgrimd Dec. 
4 (JcJ, 2 $9 Satansplatz Dec. 
The scale-rows in this series vary from 36 to 42. The majority have 38, but 
one male from Rehoboth has 40 and a female from Windhoek 42. This extends 
the range of variation to a sufficient extent to include Mabuya ansorgii Boul. of 
Benguela ; but it is possible that the latter may be a tenable subspecies, for adults 
retain the six-lined colour pattern, which is the juvenile livery in Damaraland. 
Werner (1910), in discussing a series from Damaraland, refers to a var. sexstriata 
which he ascribes to Bocage, The latter author described the coloration of the 
six-lined variety, but applied no varietal name ; consequently, the name sexstriata 
must date from Werner 1910, but it does not appear to be a valid subspecies and 
must be placed in the synonomy. In the same paper the author mentions another 
NOYITATES ZOOLOOICAE XL. 1930. 137 
example which he says is " really oclslriata " ; the word is placed in inverted 
oommas and is obviously being used descriptively and not nomenclatorially. 
60 ? Mabuya damarana (Peters, 1869) (text-fig. 43). 
Euprepes damaranius Peters, 1869, Oefrtrv. Vet.-Ak. Forhandh^ p. 660. 
Mahuia hildehrandti (non Peters) Werner, 1910, Jena Denk., xvi, p. 347 ; Boulengcr, 1910, Ann. S. 
Afr. Mus., V, p. 485; Sternfeld, 1911. Mill. Zool. Mits. Berlin, v, 3. p. 408; idem, 1911, Fauna 
Devlsch. Kolon., iv, 2, p. 40. 
Mahuia varia var. longiloha, Methuen and Hewitt. 1914, Ann. Tranxvaal Mux., iv. p. 142. 
4 Hoifnung Dec. -Jan. 
3 Windhoek Jan. 
2 Maltahiihe Dec. 
2 Voigtsgrund Dec. 
Werner, in first recording the north-east African M. kUdebrandti from 
Damaraland and S. Africa (191(», p. 347), suggested that his specimens might not 
be correctly identified, but might represent a distinct species. Methuen and 
Hewitt realized that specimens from this region, which were probably identical 
with the so-called hUdebrandli of Werner, Boulenger and Sternfeld, were more 
closely allied to varia and proposed a new subspecific name for them. The 
material in the present collection confirms the opinion that this S.W. African 
skink is not related to MMebrandti, from which it may be distinguished by the 
smaller scales on the soles of the feet and nnich .•shorter digits nnil chiws. It 
certainly seems closely akin to varia, but, in addition to the length of the ear- 
lobules, which was the only constant character discovered by Methuen and 
Hewitt to distinguish the two, there is also a constant difference in the jxisition 
of the nostril which, in all tiie material examined, affords a clear-cut differentia- 
tion between the two, without any sign of intergradation. Conse(£uently it is 
proposed to accord full specific status to the south-western form, for which, on 
geographical grounds, the name damuruna ajjpcar.s to be available in ])rcfcrcnce 
to the much later longiloha of Methui'U and Hewitt. 
M. varia and M. damarana may be distingui.shed thus : 
A. Nostril lateral, separated from the first upper labial by a distance nuich 
less than its own diameter (text-fig. 42), its centre behind the rostro-labial suture. 
Subocular not, as a rule, very much narrowed infoiiorly, its labial margin half, 
or more than half, the lengtii of its iijipcr border ; ear lobules short and broad. — 
M. varia. 
B. Nostril directed vertically ii|)\vards, sei)arated from the first u|>|irr labial 
138 NOVITATES ZOOLOGICAE XL. 193fi. 
by a distance about as great as its own diameter (text-fig. 43), its centre vertically 
above the rostro-labial suture. Subocular usually more narrowed inferiorly, its 
labial margin not half the length of its upper border ; ear lobules long and 
lanceolate. — M. damarana. 
The character of the subocular is by no means definitive and the position 
of the centre of the nostril relative to the rostro-labial suture also shows some 
variation. M. varia has been examined from the Cape Province, Transvaal, 
Natal, Angola, N. Bechuanaland, S. Rhodesia, N. Rhodesia, Nyasaland, Mozam- 
bique, Tanganyika Territory, Kenya Colony, Uganda and Somaliland. 31. 
damarana appears to be confined to the south-western districts, material having 
been seen from Damaraland, Namaqualand (Narudas Siid) and the Cape Province 
(Deelfontein and Port Elizabeth) ; Methuen and Hewitt record it from Great 
Namaqualand and various localities in the Karoo (SteythervUle, Victoria West, 
Middleburg, Klerksdale, Cradock and Steinkop). To prevent future confusion 
it must be pointed out that Fitzsimon's (1935, p. 369) inclusion of Methuen and 
Hewitt's Great Namaqualand records under M. varia varia is erroneous ; these 
examples were the cotypes of longiloba. 
61. Mabuya varia (Peters, 1867) (text-fig. 42). 
I Mt. Moco March 
In specimens of this species from Angola and the Lower Congo, the supra- 
nasals tend to be separated. In 9 out of the 10 specimens examined the fronto- 
nasal just touches the rostral. This condition seems to be rare, though not 
unknown, in specimens from other areas, and its frequent occurrence in this 
western area may indicate the beginnings of subspecific differentiation. 
62. Mabuya acutilabris (Peters, 1862). 
1 Swakopmund Feb. 
1 Voigtsgrund Dec. 
4 Lobito, Angola March 
The example from Voigtsgrund has 34 scale-rows at the middle of the body 
and exhibits an anomalous fusion of the praefrontals and frontonasal. 
63. Mabuya binotata (Bocage, 1867). 
2 near Sissekab Nov. 
This appears to be the first record of this Angolan species from Damaraland, 
but as has already been pointed out ( Nucras intertexta damarana, q.v.), Sissekab, 
with open forest country, is very different from most of the rest of Damaraland. 
64. Mabuya raddoni (Gray, 1845). 
I I Congulu AprU 
65. Mabuya maculilabris (Gray, 1845). 
6 Congulu AprU 
2 Quirimbo May 
NOVITATES ZooLonicAE XL. 1936. 139 
66. Lygosoma dewittei (Loveridge, 1934). 
1 ad. Congulu April 
This single specimen appears to agree with the species described by de 
Witte and subsequently renamed by Loveridge. One of the characteristic 
features which was believed to distinguish the species was a laterally compressed 
tail with a series of transversely enlarged subcaudals ; unfortunately the tail 
of the Congulu specimen is incomplete, but compressed tails have been noted in 
many other skinks and enlarged subcaudals appear frequently on regenerated 
tails. The species was originally referred to Siaphos, but this is now included 
by Smith (1935, p. 279) in Lygosoma. 
67. Ablepharus cabindae Bocage, 1866. 
4 Congulu April 
Smith (1935, p. 309) has recently drawn attention to the fact that the genus 
Ablepharus is not a natural one, but a polyphyletic assemblage, and has also 
pointed out that the lower eyelid is not always fused to the upper completely. 
The present species illustrates both ofthe.se facts. The lower eyelid is not fused 
with the upper except at the corners. In this character, in the presence of supra- 
nasals, and the occasional presence of four supraoculars, it differs from all the 
other African species grouped with it, and there seems to be every probability 
that it is closely allied to the West African species grouped under Riopa 
(i.e. brevicep.s, togoense, kitsoni and daJiomeyense). These species have probably 
nothing whatever to do with true Riopa and, if they be grouped with cabindae, 
as seems logical, the name Panaspis Cope (Type species P. aeneus = cabiiidae) 
becomes available for them. 
The species is variable in other characters. Thus, of four specimens here 
recorded, one has four supraoculars, whereas in the other three the first and second 
are fused, and one of the latter has the frontoparietals and interj)arietal fused to 
form a single large shield ; scales about the middle of the body vary from 24 to 26. 
68. Lygosoma (Panaspis) breviceps (Peters, 1873). 
5 Congulu April 
The discovery of this skink in Angola is a considerable extension of its known 
range. The specimens do not appear to differ from those found in the Cameroon- 
Gaboon area, though the number of scales about the body may be somewhat 
lower. They vary from 30 to 34, the known range of breviceps (including balesi 
Boul.) is from 32 to 38 (Miiller, 1910, p. 588). 
69. Riopa sundevallii (Smith, 1849). 
2 Ukahandja Dec. 
7(1. Acontias meleagris (Linn., 1758). 
5 Okahandja Dec. -Feb. 
71. Feylinia currori Gray, 1845. 
1 juv. Congulu April 
140 
NOVITATES ZOOLOOICAE XL. 1936. 
72. Amphisbaena quadrifrons Peters, 1862. 
6 HofFiiiing Dec. -Feb. 
1 Okahandja Feb. 
Cott (1933, p. 160) records 7 examples of this species from Mozambique, and 
at the same time points out that they differ from typical examples from Damara- 
land in having fewer segments in an annulus, fewer annuU on the body, 6 instead 
of 4 preanal segments and certain differences in the head-shields. The compara- 
tive material available at that time was very small, but with the new material 
in the present collection, the records of Fitzsimons (1935, p. 353) and the material 
in the Congo Museum recorded by de Witte (1933, p. 72) it becomes apparent that 
the variation is continuous across the continent. The number of preanal seg- 
ments and the variation in the head-shields appear to have no significance, and 
the numerical variations may be tabulated thus : 
73. Monopeltis capensis Smith, 1849. 
1 Rehoboth Dec. 
2 ? 
These three specimens show some variation beyond that previously recorded 
for the species. There are constantly only two maxillary teeth ; the number of 
segments in an annulus is variable, even in closely adjacent annuli, but the limits 
appear to be between .51 and 56. The annuli on the body are 203, 209 and 211, 
and the caudals constantly 11 ; preanal pores are only feebly indicated. With- 
out a larger series for purposes of comparison the significance of the smaller 
segments and reduced number of maxillary teeth cannot be ascertained. 
74. Chamaeleo etiennei Schmidt, 1919. 
5 $$, 1 juv. Congulu April 
Schmidt (1919, p. 574) drew attention to the fact that the so-called 
Ch. gracilis of the Lower Congo had a diflFerently shaped casque from typical 
W. African examples, and that, in addition, the male lacked a tarsal spur ; he 
further suggested that the Angolan specimens generally referred to gracilis would 
prove to belong to the Lower Congo form, and this suggestion is well founded. 
But the material examined shows that this species has an even wider range, as may 
be seen from the following lists : 
C. etiennei. 94 sjiecimens examined from Angola (Congulu, Pungo Andongo, 
Duque de Bragan9a, Condo, Canhoca, Marimba) and the southern half of the 
NOVITATES ZOOLOOICAE XL. 1936. 141 
Belgian Congo (Lower Congo, Banana, Zambi, Leopoldville, Vista, Congo da 
Lemba, Moanda, Boma, Kisontu, Kwango, Luabo, Mwanza Kulu and Albert- 
ville) from the coast to Lake Tanganyika. 
C. gracilis. 84 specimens examined from Gambia, Liberia, Sierra Leone, 
Gold Coast, Ashanti, Nigeria, Northern Belgian Congo (Uele, Ituri and Lake 
Albert areas), Anglo-Egyptian Sudan, Ethiopia, Uganda, Kenya and Italian 
Somaliland. 
75. Chamaeleo dilepis Leach, 1819. 
This series contains representatives of two distinct forms of this very variable 
species. Those from the forested Mt. Moco are very much smaller {,^^ 84-102, 
9$ 94-104 mm. from snout to vent) with a flatter head, less concave interorbit, 
less marked parietal crest and slightly larger occipital lobes. But until a survey 
of all tlie local races can be undertaken the use of trinomials seems inadvisable. 
AMPHIBIA SALIENTIA. 
76. Rana albolabris acutirostris subsp. n. 
$ Congulu Hplotype. 
4(?c?i ? Quirimbo Paratjrpes. 
These, the most southerly recorded examples of Rana albolabris, differ from 
a large series ranging from Liberia eastwards to Uganda and south to the mouth 
of the River Congo, in their much more acutely pointed and prominent snouts. 
This probably indicates the existence of a southern race, rather than a distinct 
species, but this is so clearly marked as to justify the use of a trinomial. A 
detailed description is not necessary, for in almost all respects, except the shape 
of the snout, these 6 specimens agree with typical albolabris. The snout is, how- 
ever, acutely pointed and very strongly prominent, wth a more obtuse canthus 
rostralis and more oblique loreal region. The distance between the nostril and 
the tip of the snout is contained not more than once and a quarter in the inter- 
narial distance, whereas in the typical form the same ratio is 1-5 to 1-7. The 
holotype, a mature female, measiu-es 82 mm. from snout to vent, and has greatly 
distended ovaries and enlarged oviducts ; it was captured in April. Another 
female caught in May is in the same conditi(m, and males caught in the same 
months have nuptial pads. The largest of the latter sex measures 74 mm. from 
snout to vent. 
77. Rana fuscigula angolensis Bocage, 1866. 
G (SS, ^> ??. 1 juv. Quirimbo May 
5 (JcJ. 5 ??, 2 juvs. Congulu April 
12 (JcJ, 3 ??, 13 juvs. Mt. Moco Jlarch 
78. Rana fuscigula fuscigula Duni. and Bibr., 1841. 
6 (J<J, 5 $$ Biillsport, Naukluft Mts. Dec. 
142 NOVITATES ZooLOOICiVE XL. 1936. 
79. Rana (Ptychadaena) ansorgii Boulenger, 1905. 
2<J^, $ Mt. Moco March 
cj Congulu April 
It will probably be found that aU records of " ansorgii " from Tanganyika 
Territory, Portuguese East Africa, Nyasaland, Rhodesia and Zululand really 
refer to a distinct species, which is similar in its digital webbing, but has a broader 
head and less prominent snout, characters in which it resembles R. mascareniensis. 
The name available for this eastern form appears to be mossambica Peters, 1854. 
80. Rana (Ptychadaena) oxyrhynchus Smith, 1849. 
$, juv. Congulu April 
1 juv. 
13 6<S 
82. Pyxicephalus delalandii Dum. and Bibr., 1841. 
$ Windhoek Jan. 
8 (^(J, 12 ?$ Farm Hoffnung, near Windhoek Oct.-Jan. 
2 juvs. Omongongua Jan. 
2 juvs. Okahandja Feb. 
19 (J (J, 2 $$ Maltahohe Dec. 
This species was found breeding in December and January. Dr. Jordan 
reports that the voice of the males is loud but melodious, sounding at a distance 
like " lutter, lutter, lutter " (the u sound being long) ; it is accordingly 
known locally as the " Lutheran." 
83. Pyxicephalus adspersus Tschudi, 1840. 
1 juv. Omongongua January 
84. Phrynobatrachus natalensis (Smith, 1849). 
7 SS'"' ?? Congulu April 
85. Arthroleptis parvulus Boul., 1905. 
34 c?(?. 4 ?? Calaongo, below Congulu April 
This series shows two features which have not previously been recorded for 
the species. First of all, males have a large, flat, oval gland on the hinder side of 
the thighs as in A. cornutus, and A. ogoensis (Parker, 1935, p. 403). Hitherto 
parvulus has not been suspected of relationship with these species, but has been 
grouped with A. dispar and A. feae (Noble, 1924, p. 201) ; examination of the 
latter two species shows that these femoral glands are present in the males of 
dispar, but not, apparently, of feae. 
Secondly there appears to be a considerable amount of variation in the degree 
of webbing of the toes. The majority of individuals have the toes about a quarter 
webbed, as in the cotypes, but two others in the present collection, and two 
NOVITATES ZOOLOGICAE XL. 1936. 143 
others in the British Museum from Mossamedes, have them nearly half webbed, 
as in ogoensis ; they differ from this species, however, in their nnich smaller 
digital discs. It seems very improbable that these four specimens represent a 
distinct species, for they differ in no other characters and even in more normal 
specimens there is no absolute constancy in the webbing. The resemblance of 
this species to dispar is very marked, but the latter has larger digital discs as in 
ogoensis ; nevertheless it seems very probable that the two may have been con- 
fused and that the Angolan records of " dispar " (Peters, 1888, p. 618) really refer 
to parvulus. 
86. Cacosternum boettgeri (Boul., 1882). 
14(;J(J Hoffnung, near Windhoek Dec. 
c? „ „ „ Feb. 
(> SS Maltahohe Dec. 
(J Voigtsgrund Dec. 
This species was breeding at Hoffnung in December ; the voice of the male 
resembles castanets. 
87. Phrynomerus annectens (Werner, 1910). 
Phrynmnanlia nusiUus Methuen and Hewitt, 1914, Ann. Transvaal Mvs., iv, p. 122, pi. xiv, fig. 2. 
Hoplophryne marmorata Ahl, 1934, Zool. Anz., cvii, p. 334, fig. 1. 
2 juvs. Morro de Pundo May 
These two specimens both have a somewhat longer leg than has been 
described previously, and as all the previous records appear to have been from 
S. Africa and the Cape Province, this might suggest the existence of a distinct 
northern race in which the tarso-metatarsal articulation nearly reaches the eye 
(instead of the shoulder). But the difference is sexual, for out of a series of 8 
specimens from Benguela the 4 males have the tarso-metatarsal articulation 
reaching the posterior corner of the eye, whilst in the four females it only reaches 
the axilla or shoulder. 
88. Kassina senegalensis (Dum. & Bibr., 1841). 
Kasaina deserlicola Ahl, 193U, Zool. Anz., Ixxxviii, p. 280 ; idem, 1931, jDa« Tierreich, Aniira, iii, p. 449. 
3 c?cJ Hoffnung Dec. 
3 (Jc? .. Jan. 
This series, which is almost topotypical of K. deserlicola, shows a consideral)le 
variation in the degree of folding and wartiness of the anterior part of the belly ; 
other specimens in the British Museum from other localities also show con- 
siderable variation, and since this is the principal feature said to distinguish 
deserlicola from senegalensis it seems very doubtful whether the former can bo 
retained, even subspecifically. 
The species was breeding in December. Males do not call from the water, 
but from beneath stones or herbage not far from it. The call is a short guttural 
" ou-i " (not unlike the sound of vomiting) and is not given continuously, but at 
long and irregular intervals. Power luus described the mating call as rcsenihlirig 
the withdrawal of a cork from a bottle or the bursting of a large bubble (1925). 
89. Hyperolius bocagei Steindachner, 1869. 
cJ, $ Congulu AprU 
144 NoVITATES ZOOLOGICAE XL. 1936. 
90. Hyperolius cinnamomeoventris Socage, 1866. 
1 $, 1 juv. Congulu April 
These two specimens agree well in morphological characters with repre- 
sentative specimens of the species. In the female, however, the characteristic 
black lateral markings are only very faintly indicated, whilst in the juvenile they 
are completely absent. 
91. Hyperolius sp. 
$ Mt. Moco March 
This single specimen belongs to a very short-webbed, uniformly-coloured 
species which may well be undescribed. 
92. Leptopelis jordani sp. n. 
Holotype a female from Congulu, 700-800 m. ; collected in April 1934 by 
Dr. Karl Jordan. 
Vomerine teeth in two groups between the choanae. Head broad, very 
much broader than long, with a blunt snout once and a quarter as long as the 
eye ; canthus rostralis obtusely angular, strongly curved ; loreal region oblique, 
very slightly concave. Tympanum distinct, slightly more than half the diameter 
of the eye and separated from the latter by a distance greater than half its own 
diameter. Digits with well-developed discs and strong subarticular tubercles. 
Fingers with a rudiment of web ; first shorter than the second, which is shorter 
than the fourth. Toes not quite half webbed. A very large compressed, inner 
metatarsal tubercle as long as its distance from the disc of the inner toe. Limbs 
short, the tibio-tarsal articulation reaching the shoulder and the metatarsal shovel 
the anterior corner of the eye ; tibia not three times as long as wide, its length 
equal to the maximum width of the head and contained 2-5 times in the length 
from snout to vent. Skin smooth above ; a few granules below the ear. Lower 
surfaces, including the thighs, strongly granular. 
Purplish brown above ; a black line from the tip of the snout, through the 
nostril and eye, along the upper margin of the tympanum to the flanks. An 
obscure, dark, interorbital triangle connected by its apex to a dark chevron on 
the middle of the back ; posterior part of the back dark-stippled. A series of 
white streaks forms lines along the outer edge of the forearm, the outer edges of 
the tarsus and metatarsus and above the vent. Lower surfaces yellowish white, 
a few spots of this coloiu' invading the flanks. Chin faintly brown-mottled. 
Length from snout to vent 62 mm. ; width of head 24 mm. ; hind-limb 
85 mm. 
This species is undoubtedly the representative in the Congulu Forest-zone 
of the widespread L. aubryi of the Rain Forest proper. It differs from the latter 
in its shorter leg, broader head, larger size and the greater distance intervening 
between tympanum and eye. In a series of 27 examples of L. aubryi from French 
Guinea to the Ituri and south to Gaboon the tibio-tarsal articulation always 
reaches the eye and the metatarsal tubercle past the tip of the snout ; the tibial 
length is always appreciably greater than the width of the head and contained 
only 2-2-25 times in the length from snout to vent ; the greatest length is 54 mm. 
and the distance between eye and tympanum less than half the diameter of the 
latter, though subject to some variation. 
NOVITATES ZOOLOGICAE XL. 1930. 145 
93. Bufo jordani sp. n. 
Holotype a j', from Satansplatz, circa 1,3(10 ra. ; collected Dec. 1(134 by 
Dr. Karl Joiclan. 
Crown without bony ridges ; snout bluntly rounded, once and a quarter as 
long as the eye ; canthus rostralis rounded but distinct ; loreal region nearly 
vertical ; tympanum and eustachian tubes absent ; interorbital space flat 
and broader than tlie upper eyelid. Fingers short, the first shorter than the 
second, which is shorter than the fourth ; third finger, measured along its mesial 
side, as long as the snout ; two large, flat, metacarpal tubercles. Toes nearly 
one-third webbed, with double subarticular tubercles ; sole with conical tubercles ; 
two metatarsal tubercles ; no tarsal fold ; tarso-metatarsal articulation reaching 
the eye and the length of the tibia contained 2-8 times in the length from snout 
to vent. 
Paratoid glands absent. Dorsal surfaces closely beset with small conical 
warts each of which is tipped with a minute, blunt spine (J). Lower surfaces 
wrinkled, but not granular. 
Uniform purplish brown above ; immaculate white beneath. 
Length from snout to vent 28 mm. 
Nuptial asperities are developed on the inner two fingers ; vocal sacs 
absent. 
This species is allied to B. anotis Boulenger and B. katanganus Loveridge, 
both of which it resembles in its reduced auditory apparatus ; it is distinguished 
from both by the absence of paratoid glands and by its ungranulated lower 
surfaces. These species, and also B. taitaniis Peters, B. rosii Hewitt, B. lonn- 
bergi Andersson, B. mocqiiardi Angel, B. preussi Matschie, B. surdus Boul., 
B. fissipes Boul., B. variegatus (Giinther) and B. ockendeni Boul., have the tym- 
panum not merely hidden but absent ; there is no cuvurn typaiii, annulus lym- 
panicus, columella auris or Eustachian tubes as in the many genera previously 
recorded by the author (1934, p. 4). The same condition probably obtains in 
B. micranotis Loveridge, B. ushoranus Loveridge and B. osgoodi Loveridge. 
94. Bufo regularis Reuss, 1834. 
11 (J(J, 2 $9 Otjosongombe Nov. 
juv. Sissekab „ 
3 (^O, 6 juvs. Cuito, Mt. Moco March 
5 Congulu April 
95. Xenopus laevis laevis (Daud., 1803). 
12 SS< ?? Otjosongombe Nov. 
2 ?$ and late larva Hoflfnung 28-31 Dec. 
5 Okahaiidja Oct. 
1 1 larvae and recently 
metamorphosed exs. Voigtsgrund 14 Dec. 
Most of the.se s])ccimens are quite typical, with immaculate lower surfaces, 
tliougli generally small. The single female from Okahandja, however, is very 
much larger (92 mm.), but hits the mottled abdomen more commonly found in 
the Angolan peterbi (Parker, 1936). 
10 
146 NOVTTATES ZoOLOaiCAE XL. 1936. 
Ahl, 1924, Archiv Nalurg., xc, A. 5, p. 247. 
Angel. 1923, Miss. Rohan-Chahot Angola d- Rhodesia, Rept., pp. 1-13, pi. 
Socage, 1895, Herp. Angola and Congo. 
Boettger. 1887. Ber. Senck. Nat. Ges. 1886-1887, pp. 55-64. 
Boettger, 1887, Ber. Senck. Nal. Ges. 1886-1887, pp. 135-173, pi. v. 
Boettger, 1894, Ber. Senck. Nat. Ges. 1894, pp. 83-93. 
Boulenger, 1893, Cat. Snakes Brit. Uus. I. 
Boulenger, 1894, Cat. Snakes Brit. Ahis. II. 
Boulenger, 1896, Cat. Snakes Brit. Mus. III. 
Boulenger, 1907, Mem. Proc. Manchester Litt. Phil. Soc., li, 3, no. 12. 
Boulenger, 1910, Ann. S. Afr. Mus., v, 9, pp. 455-538. 
Boulenger and Power, 1921, Trans. R. Soc. S. Africa, ix, 3, pp. 229-287. 
Boulenger and Power, 1921, Trans. R. Soc. S. Africa, x, 1, Map. 
Cott 1934, Proc. Zool. Soc. London, pp. 145-173, pis. I-III. 
Fitzsimons, 1935, Ann. Transvaal Mus., xvi, 2, pp. 295-397, pis. x-xi. 
Hewitt, 1910, Ann. Transvaal Mus., ii, pp. 73-115. 
Hewitt, 1926, Ann. S. Afr. Mus., xx, 6, pp. 413^31 , pi. xxxvii. 
Hewitt, 1927. Rec. Albany Mus., iii, pp. 371-415, pis. xx-xxiv. 
Hewitt, 1932, Ann. Natal Mus., vii, 1, pp. 105-128, pi. vi. 
Hewitt, 1935. Rec. Alhany Mus., iv, pp. 283-357, pl.s. xxvii-xxxvi. 
Loveridge, 1932, Proc. Biol. Soc. Washington, xlv, pp. 83-86. 
Loveridge, 1932, Bull. Mus. Comp. Zool., Harvard, Ixxii, 10, pp. 375-387. 
Loveridge, 1933, Bull. Mus. Comp. Zool., Harvard, Ixxiv, 7, pp. 197-416, pis. i-iii. 
Methuen and Hewitt, 1913, Ann. Transvaal Mus., iv, 3, pp. 118-167, pi. xiv. 
MuUer, L., 1910, Ahh. Bayer. Ak. Wiss. (2), xxiv, pp. 545-626. 
Noble, 1924, Bull. Amer. Mus. N. H., xlix, pp. 147-347. 
Parker, 1932, Proc. Zool. Soc. London, pp. 335-367. 
Parker, 1933, Ann. Mag. N. H. (10), xii, pp. 544-548. 
Parker, 1934, Monograph Frogs Family Microhylidae. 
Parker, 1935, Ann. Mag. Nat. Hist. (10), xvi, pp. 401-404. 
Parker, 1936, Ann. Mag. Nat. Hist, (in press). 
Peters, 1854, Moti. Ak. Berlin, p. 621. 
Peters, 1865, Mon. Ak. Berlin, p. 261, fig. 5. 
Peters, 1877, Mon. Ak. Berlin, pp. 611-620. 
Pitman, 1934, Rept. Faunal Survey N. Rhodesia (Livingstone). 
Power, 1925, Trans. R. Soc. S. Africa, xiii, p. 108. 
Power, 1930, Ann. Transvaal Mus., xiv, 1, pp. 11-19, pis. i-ii. 
Roux, 1907, Zool. Jahrb., Syst., xxv, pp. 403-444, pis. xiv-xv. 
Sanderson, 1936, Proc. Zool. Soc. London, pp. 165-208, PI. 1. 
Schmidt, 1919, Bull. Amer. Mus. N. H., xxxix, 2, pp. 385-624. 
Schmidt, 1923, Bull. Amer. Mus. N. H., xlix, pp. 1-146. 
Schmidt, 1933, Ann. Carnegie Mus., xxii, pp. 1-15, pis. i-ii. 
Sjostedt, 1897, Bihang Si'enska Vet.-Ak. Handl., xxiii, 4, ii. 
Smith, M. A., 1935, Fauna Brit. India, Rept. dk Aviph., II. 
Steindachner, 1882, Sitz. ber. Ak. Wiss. Wien., Ixxxvi, 1, pp. 1-3, pi. 
Sternfeld, 1908, Sitz. ber. Ges. Natf. Fr. Berlin, i, pp. 92-95. 
Sternfeld, 1910, Fauna Deutsch. Kolon., iv, 1. 
Sternfeld, 1911, Fauna Deutsch. Kolon., iv, 2. 
Sternfeld, 1911. Mitt. Zool. Mus. Berlin, v, 3, pp. 395^1 1 . 
Sternfeld, 1912, Wi.<is. Ergeb. Deutsch. Z. Afr. Exp., ix, Zool. 2, pp. 197-279 
Thominot, 1878, Bull. Soc. Philo^m. Paris, pp. 1-3. 
Werner, 1909, Milt. Nat. Mus. Hamburg, xxvi, pp. 205-247. 
Werner, 1910, Jena Denkschr., xvi, pp. 279-370, pis. vi-xi. 
Werner, 1913, Denks. Ak. Wiss. Wien, Ixxxviii, pp. 1-6, pi. 
de Witte, 1933, Ann. Mus. Congo Beige, Zool. (1), III, 2. 
de Witte, 1934, Ann. Mus. Congo Beige, Zool. (1), III, 4.