550 



NA TURE 



[September 27, 1906 



ing, and is characterised by an increasing frequency of 

 Marl bands. Tfiese he regards as evidences of shallow 

 lakes, and compares them with similar pools now found 

 in the desert regions of South Africa. The muds forming 

 the floors of these pools, both in the recent and older 

 examples, contain Estheria, and afford impressions of foot- 

 prints, raindrops, and desiccation cracks. The Keuper 

 .Marls he compares with the Loess of eastern Europe, and 

 the beds of salt, gypsum, and other salts he regards as the 

 result of evaporation in lakes. 



Mr. Holland referred to certain phenomena in the 

 Kajputana desert that supported Mr. Lomas's views with 

 regard to the processes of concentration in arid regions, 

 and gave evidence of the sifting action of wind in India. 

 Similar bands of silt and mud are found filling in hollows 

 in the ArchjEan rocks. He was not prepared to admit 

 lliat the features of the British Trias were due only to 

 wind action, but in the main they were due to conditions 

 prevailing in desert regions. Prof. Cole pointed out that, 

 in dealing with the British Trias, we must not forget the 

 great sea eastwards and the likelihood of the establish- 

 ment of a monsoon system on its margin. This might set 

 up an intense rainy season for, say, three months in the 

 year, followed by a dry season. Sheets of pebbles without 

 well-defined water channels are compatible with general 

 evidence of desiccation. Mr. R. D. Oldham showed thqt 

 the only agency forming pure sands comparable with the 

 Trias is wind. Mr. Clement Reid compared the peculiar 

 stiff-stemmed flora of the desert with those found in the 

 Trias. 



The papers dealing with pateontology were more than 

 usually interesting. Mr. C. G. Danford exhibited and de- 

 scribed a fine series of ammonites from Speeton. Mr. 

 A. C. Seward dealt with the Jurassic flora of Yorkshire, 

 and Dr. H. Woodward, in describing a wonderful collec- 

 tion of arthropods from the Coal-measures at Sparth 

 Bottoms, showed what an enthusiastic band of collectors 

 can do, when work is taken in hand in the spirit which 

 characterises the Rochdale geologists. 



The report on the fauna and flora of the Trias included 

 an important paper by Dr. \. Smith Woodward on 

 Rhynchosaurus articeps, and Mr. H. C. Beasley and Mr. 

 Lomas described the great finds of Triassic foot-prints 

 which have recently been discovered at Storeton, in 

 Cheshire, and Hollington, in Staffordshire. 



In petrology and mineralogy great interest was shown 

 in the announcement by Prof. Edgeworth David that 

 diamonds had been found embedded in the matrix near 

 Inverell, New .South Wales. 



Mr. T. H. Holland demonstrated the peculiar properties 

 of a variety of sodalite from Rajputana. When freshly 

 broken it has a bright carmine colour, which changes to 

 dull grey on being exposed to light. The carmine colour 

 n-turns when the specimen is kept in the dark. Prof. 

 H. S. Reynolds dealt with the igneous rocks in the district 

 south-west of Dolgelly, and described the occurrence of a 

 picrite from the eastern Mendips. 



In general geology Mr. J. Parkinson gave an interest- 

 ing account of the post-Cretaceous geology of Southern 

 Nigeria, Prof. Cole outlined a scheme of geology suited 

 to agricultural scholars, and Prof. J. Milne discussed 

 certain earthquake relationships. 



While the time of the section was fully taken up by 

 the consideration of the above subjects, no less than thirty- 

 seven papers dealing with strictly geological matters were 

 read in other sections. J. L. 



ZOOLOGY AT THE BRITISH ASSOCIATION. 

 'T'HE large attendance at many of the meetings of 

 Section D was sufficient evidence of the general 

 interest of the programme, which included discussions upon 

 the Tanganyika problem, the nature of fertilisation, spicule 

 formation in sponges, the bearing of scientific marine in- 

 vestigations on practical fishery problems, and a number of 

 papers on special subjects, only a few of which can be 

 noticed here. 



The Tanganyika Problem. 



The discussion on the Tanganyika problem was opened 



by Mr. J. E. S. Moore, who dealt, first, with the 



characters of fresh-water faunas in general, pointing out 



NO. 1926, VOL. 74] 



the wide distribution of many fresh-water organisms over 

 the land surfaces of the world. He held that the difficul- 

 ties in the way of the migration of these animals were 

 so great that their wide distribution could not be attributed 

 solely to such migration. He suggested that in all prob- 

 ability the sea is becoming more salt, and that this change 

 may have been concerned in the production and separation 

 of marine and fresh-water faunas. Whatever the actual 

 cause of separation, as the general fresh-water fauna of 

 the globe possessed certain archaic characters it would be 

 convenient to name this the primary fresh-water fauna. 

 To this primary fauna there are added in many places, 

 e.g. in the Caspian Sea, animals which have, from their 

 structure and alifinities, been obviously derived from the sea, 

 and have an origin independent of that of the fresh-water 

 fauna of the region in which they occur. To these animals 

 Mr. Moore applied the name halolimnic. There are in 

 Tanganyika a number of animals peculiar to that lake, and 

 regarded by Mr. Moore as halolimnic. The mollusca of the 

 lake are represented by certain ordinary fresh-water forms, 

 but, in addition, there are several not closely related to any 

 recognised fresh-water type, nor does their anatomy suggest 

 that they have been evolved from any African fresh-water 

 form ; there are four Polyzoa, only one of which is 

 phylactolaematous, and it may be inferred that the other 

 three are derived from marine forms, while the occurrence 

 of a medusa is also suggestive in this connection. There 

 are three possible explanations of these faunistic peculiari- 

 ties : — (i) that they are due to direct modifications of the 

 general African fresh-water fauna ; (2) that they are con- 

 stituted by the presence in the lake of the remains of an 

 extinct fresh-water fauna ; (3) that they are due to the 

 presence of halolimnic elements. Mr. Moore regarded the 

 last as the correct explanation, and referred to the 

 similarity of the shells of certain Tanganyika gastropods 

 to those common in Jurassic seas. The evidence points to 

 Tanganyika having been isolated a long time from the 

 sea. 



Mr. W. A. Cunnington gave a brief account of the third 

 Tanganyika expedition, from which he had recently re- 

 turned. 115 fishes are now recorded from the lake, 102 of 

 which occur nowhere else. Twelve species of prawns (of 

 which only one has been found elsewhere) are all 

 specialised in the direction of reduction of gills, and the 

 four species of crabs are all endemic. These facts are 

 probably to be explained by the long isolation of the lake. 

 It is curious that no Cladocera were met with in Tangan- 

 yika, though they are abundant in Victoria Nyanza and 

 Lake Nyassa. 



Prof. J. W. Gregory considered that there are no 

 evidences of marine rocks in the plateau of equatorial 

 Africa, though it is evident that the plateau is of great 

 antiquity. The idea of the occurrence of the sea in the 

 Tanganyika valley should be abandoned. He suggested 

 that the " halolimnic " fauna is rather to be explained as 

 a part of an ancient lake fauna at one time widely dis- 

 tributed over Africa, but now surviving only in Tanganyika. 



Prof. Pelseneer pointed out that the external resemblances 

 of shells are often illusory, and the results to which they 

 lead quite uncertain, therefore only the study and com- 

 parison of the internal organisation of the molluscs can 

 throw light on the question at issue. Messrs. Moore and 

 Digby have suggested that some of the Tanganyika 

 molluscs have affinities to certain marine forms, Chytra 

 being related to Hipponyx and Capulus, Spekia to 

 Lamellaria, and Edgaria ( = Nassopsis) to the Architaenio- 

 glossa ; but Prof. Pelseneer held that there are really no 

 affinities, in the usual sense of the word, between these 

 forms, but only distant resemblances, such as are common 

 to all the Tfenioglossa, to which group these " halo- 

 limnic " forms belong. Nor do they present archaic 

 characters to a greater extent than other fresh-water genera 

 not " halolimnic," such as Ampullaria and Paludina. Prof. 

 Pelseneer concluded that all the " halolimnic " gastropods 

 belong to the family Melaniidae or to closely related types, 

 as is shown by their radulas, otocysts, &c., and by special 

 details of their biology — their fresh-water habitat and 

 viviparity. The study of two genera (Giraudia and 

 Lavigeria) the organisation of which has only just been 

 investigated supports this conclusion. Both have in their 

 otocysts multiple otoliths, one otolith being much larger 



