TRANSACTIONS OP SECTION D. 707 



than in any otlier known forms. The teeth in the allied genus Brontothenum 

 have already heeia figured and described by the author. 



The vertebrae are somewhat similar to those of the existing rhinoceros. In the 

 present genus, Bi-ontops, the neural spines of the dorsal vertebrae are elevated and 

 massive. There are four sacral vertebrpe in this genus, and in the known species 

 the tail is short and slender, as in the individual here described. 



The ribs are strong and massive. The sternal bones are compressed transversely. 

 The exact form of the first one is not known with certainty, and is here restored 

 fi-om the rhinoceros. This is the only important point left undetermined in the 

 restoration. 



The fore limbs are especially robust. The humerus has its tuberosities and 

 ridges very strongly developed, and the radius and ulna have their axes nearly 

 parallel. There are four well-developed digits in the manus, the first being entirely 

 wanting. 



The pelvis is very wide, and transversely expanded, as in the elephant. The 

 femur is long, and has the third trochanter rudimentary. The tibia and fibula are 

 quite short. The calcaneum is very long, and the astragalus is grooved above. 

 There are only three digits in the pes, the first and fifth having entirely 

 disappeared. 



Diplacodon of the Upper Eocene is clearly an immediate ancestor of the 

 Brontotheridm, while Palccosijops and Lhnnohyus of the Middle Eocene are on the 

 more remote ancestral line. The nearest related European form is the Miocene 

 Chalicotheriuin. No descendants of the Brontotheridce are kno^vu. 



Menodus, Megacerops, Brontotherium, Symborudon, Menops, Titanops, and Allops 

 all belong to the family Brontotheridce., and their relation to the genus here 

 described, and to each other, will be fully discussed in the monograph to which 

 reference has already been made. 



3. Keredity in Gain with an extra Number of Toes. 

 By B. B. PoTJLTON, M.A. 



4. On the Nature of the Geological Terrain as an inifoi-tant factor in the 

 Geographical Distribution of Animals. By Hans Gadow, M.A., Ph.D. 



The observations concern the geographical distribution of all the species of 

 Amphibia and Reptiles in the Iberian Peninsula, with special reference to Portugal. 

 The number of difl"erent species amounts to 38, whilst the localities from which 

 these species are recorded number about 800. 



The method adopted — with a view of finding out a possible interdependence 

 between geological terrain and occurrence of species — was the following : The 

 various species and localities were arranged in five groups, which latter represent 

 the most typical sorts of terrain in the Peninsula, viz. : Palaeozoic ; granite and 

 other igneous rocks ; new red sandstone ; mesozoic limestones, including the 

 cretaceous deposits ; and, lastly, tertiary terrain, the latter being overlaid with sandy 

 soil chiefly. 



The proportionate areas of these five classes of terrain were given in percentage 

 of the area of the whole country. 



The most reliable conclusions could be drawn from Portugal, since this country 

 had been extensively visited by the author on four different occasions. The follow- 

 ing conclusions refer to Portugal only; it may, however, be especially stated 

 that these results agree closely with those drawn from the whole peninsula. _ The 

 distribution of the Reptiles exhibits features diiferent from that of the Amphibia. 



Concerning Amphibian Life. — The most favourable terrain is that of red sand- 

 stone ; then follows granitic terrain, then tertiary, palaeozoic, and lastly mesozoic 

 limestone. Granitic terrain is more than six times more favourable than palaeozoic, 

 and three times more favourable than tertiary terrain. Limestone is as a rule 

 almost detrimental. Palceozoic terrain is about seven times less favourable than 

 red sandstone, and not half so good as tertiary terrain. 



z z 2 



