i6S SOME ETHIOPIAN REPTILES, FISHES, AND- INVERTEBRATES 



example of this group. The allied genus Homopus, in which somewhat similar 

 markings occur, is exclusively Ethiopian. As regards the evolution of the colour- 

 pattern on the shells of South African land-tortoises, more especially those included 

 in the genus B.omo'pus, the following suggestion has been made : Starting with 

 species in which each shield of the shell is of the normal horn-colour, the first 

 stage is the development of a dark border, followed later by a dark centre. Next 

 the whole shield becomes dark, excepting light lines radiating from the centre, 

 after which the dark area may break up into spots or flecks. 



In common with other parts of the Southern Hemisphere, equatorial and 

 southern Africa is the home of several members of the group of side-necked, 

 or pleurodiran, water-tortoises, that is to say, those which withdraw their head and 

 neck by bending it to one side, instead of retracting it in a vertical plane by means 

 of an S-like flexure. This group is unknown at the present day in the Northern 

 Hemisphere. The two genera found in Africa, namely, Sternothcerus and 

 Pelomedusa, are common to the tropical and southern portion of that continent 

 and Madagascar. Of the soft river-tortoises, the region possesses representatives 

 of the widely spread typical genus Trionyx, and also two peculiar generic types 

 Cycloderma and Cydanorbis, confined to the tropics, where the first is represented 

 by two and the second by a single species. 



Africa is the home of a large number of more or less peculiar types of lizards, 

 and from the distribution of these Mr. J. Hewitt has arrived at important con- 

 clusions with regard to the division of the continent into zoological provinces 

 After mentioning that the Zambesi-Cuneni line does not form a natural zoological 

 boundary, as there is an extensive overlap of the southern and tropical faunas, 

 the author expresses the opinion that the southern districts of Africa possess a 

 fauna sufficiently peculiar to entitle this area to be regarded as a distinct zoological 

 region, divisible into several subregions. " As regards the entity of the South 

 African region as a zoological area, there can be no doubt but that the distinction 

 between the peculiar endemic fauna of southern Africa and tropical Africa is too 

 pronounced to permit of our regarding the South African region as merely a 

 province of the large Ethiopian area, and, indeed, but for the infiltration of 

 ti-opical forms, no one would hesitate to unite South Africa with Madagascar as 

 a region quite distinct from the more northern parts of Africa, But the question 

 of the northern boundary of our area is quite another matter." Mr. Hewitt fully 

 believes in the theory of a former land-connection between Australia, India, 

 Madagascar, the Seychelles, and South Africa, which was sundered between 

 Australia and Africa late in the Secondary epoch, and was elsewhere broken up 

 into islands in the eai-ly Tertiary. The connection between Madagascar and India 

 persisted until the Eocene, or perhaps later, as an archipelago, and Africa may 

 have been connected by swamps with Madagascar until the early Pliocene or 

 later Tertiary period. Another land-bridge connecting South Africa and South 

 America by way of the Atlantic is likewise accepted. The fauna of the whole 

 area is considered to have had many features in common ; but after the separation 

 of Madagascar and the formation of the African continent the latter area was 

 invaded by a fauna from the north which could not reach Madagascar. The fauna 

 of that island accordingly seems to represent in a modernised form — with a few 



