330 



SCIENCE 



[Vol. XIX. No. 488 



and the oxalic acid, in another quantity of the salt, was de- 

 termined by means of a standard solution of potassium per- 

 manganate. The crystals of the oxalate were thus found to 

 contain 52.70 per cent of masrium oxide, 15.85 per cent of 

 oxalic anhydride, and 31.27 per cent of water. 



From the whole of the analytical data yet obtained, as- 

 suming, as the reactions of the salts would indicate, that 

 masrium is a divalent element, the atomic weight would ap- 

 pear to be 228. An element of atomic weight about 235 is, 

 indeed, required to occupy a vacant place in the periodic 

 system in the beryllium calcium group, and masrium ap- 

 pears likely to be the element in question. 



Masrium has only yet been observed to combine with 

 oxygen in one proportion, to form the oxide MsO. Masrium 

 oxide is a white substance much resembling the oxides of 

 the lime group. The chloride, MsClg, is obtained upon 

 ■evaporation of a solution of the oxide or hydrate in hydro- 

 chloric acid. The nitrate, Ms (N03)3, crystallizes from 50 

 per cent alcohol, and the crystals contain water, the amount 

 of which has not been determined. The sulphate, MSSO4 . 

 8H3O, is a white salt which crystallizes badly from water, 

 but which separates in well-developed crystals from 50 per 

 cent alcohol. It combines with sulphate of alumina to form 

 an alum, also with potassium sulphate to form a couble 

 sulphate. The oxalate above referred to, MSC3O4 . 8H3O, 

 is a white salt, soluble in acetic acid, and also in excess of 

 masrium chloride. 



The most important reactions of the salts of masrium, as 

 Jar as they have yet been studied, are the following. Sul- 

 phuretted hydrogen produces no precipitate in presence of 

 Lydrochloric acid, but yields a white precipitate in presence 

 of acetic acid. Ammonia precipitates the white hydrate of 

 masrium trom solutions of the salts; the hydrate is insoluble 

 in excess of ammonia. Ammonium sulphide and carbonate 

 produce white gelatinous precipitates, likewise insoluble in 

 excess of the reagents. Ammonium phosphate yields a 

 white precipitate of phosphate. Caustic alkalies precipitate 

 the hydrate, but the precipitate is readily soluble in excess 

 of the alkaline hydrate. Potassium ferrocyanide produces a 

 white precipitate which is soluble in excess of masrium 

 chloride, but not in dilute hydrochloric acid. Potassium 

 ferricyanide yields no precipitate. Potassium chromate 

 precipitates yellow chromate of masrium, which is soluble 

 in a further quantity of masrium chloride. Potassium tar- 

 trate yields a white tartrate precipitate whicVi dissolves in 

 excess of the reagent, but the solution is not reprecipitated 

 by the addition of ammonia. 



!Rtetallic masrium has not yet been obtained. Attempts to 

 isolate it by heating the chloride with sodium under a layer 

 of common salt, and by the electrolysis of a solution of the 

 ■cyanide proved unsucaessful. The cliloride, moreover, is 

 not sufficiently volatile to permit of its vapor density being 

 determined. 



From the above interesting reactions, however, it will be 

 evident that masrium possesses a strong individuality, 

 although on the whole behaving somewhat like the metals 

 of the alkaline earths and those of the zinc group. Further 

 work will doubtless afford more definite information con- 

 <cerning its nature and properties. A. E. Tutton 



SOME NOTES ON THE VICTORIA NYANZA. 



The following observations on the Victoria Nyanza have 

 teen sent to the Royal Geographical Society by Mr. Ernest 

 •G-edge, who has spent a considerable time on the lake and 



in its neighborhood: "The appearance of the lake suggests 

 the formation at some remote period of a vast trough or val- 

 ley; the western coasts give striking indications of this, 

 especially in Karagwe, where the cliffs come sheer down 

 with deep water close in shore. Inland, behind these, can 

 be noticed a succession of lines of fault, running parallel to 

 one another, forming a series of terraces or steps, which 

 finally culminate in the high grassy plateaus stretching 

 away westwards. There is nothing either on this side or on 

 its southern shores suggesting volcanic action ; the geologi- 

 cal structure consisting f(jr the most part of gneissic forma- 

 tions and schists, with enormous boulders of porphyritic 

 granite, the latter constituting the most prominent feature 

 on its southern coasts, as well as forming a remarkable 

 island in the lake, known as the " Makoko" or white rocks. 

 On the northern shore outcrops of honey-combed iron stone 

 and lava blocks are to be seen, and this change in the 

 geological structure is accompanied by a corresponding 

 change in the vegetation, from the sterile arid wastes so 

 characteristic of the southern coasts, to rich tropical growth. 

 The main visible sources of the water supply for this great 

 reservoir ai'e the Kagera, Nzoia, and Ngure Darash rivers; 

 and these, though continually discharging a certain amount 

 of water into the lake, are of no great size, except during 

 the rainy season, appearing totally inadequate to maintain 

 the equilibrium of the lake, when we consider the volume 

 of water constantly being carried off by the Nile,' as well as 

 the loss that must be caused by evaporation from so large an 

 area. This would lead one to suggest the existence of springs 

 to make up the deficiency. The lake is of great depth in 

 places, and the water fresh and clear, though flat and 

 insipid to drink. Fish are plentiful, being mostly caught 

 with a rod and line, the nearest approach to netting being a 

 screen of grass mats, used as a sieve by the people in Lower 

 Kavirondo, and the basket traps used by the Ba-Sesse. 

 Amongst others is a Silurus, which has evidently been mis- 

 taken for the porpoise, owing to its shiny black body, and its 

 habit of coming to the surface and indulging in porpoise- 

 like gambols in calm weather. Hippopotami are not very 

 plentiful, as they chiefly confine themselves to the coasts 

 and rivers. Those that are found in the open water are, 

 however, extremely vicious and much feared by the Ba- 

 Sesse canoe-men, who, strange to say, are unable to swim. 

 This is no doubt largely due to the fact of the lake being in- 

 fested with alligators, rendering it dangerous for any one to 

 enter the water. Cyclonic storms of great violence occur at 

 certain seasons, and are most dangerous to small craft. 

 These storms in August usually occur at daybreak, coming 

 from the south-west, with much thunder and lightning. 

 Following the coast-line for a time, they would suddenly 

 sweep across the lake in a northeast direction, raising a 

 tremendous sea, and on several occasions we were in immi- 

 nent danger of being swamped. During this month I no- 

 ticed that about 3 a.m. the wind was invariably off-shore, 

 varying from the north-north-east to north and north-west. 

 This would drop about 11 A.M., to be followed by a calm 

 lasting to about 2 P M., when the wind would again come 

 up and blow strongly, in gradually increasing force, from 

 the south west to south, dying away again at night about 8 

 P.M. During November the prevailing wind was from the 

 north-east. One of the most remarkable phenomena I wit- 

 nessed was the apparent tide observable at irregular inter- 

 vals, the waves coming in and overflowing the beach in 

 exactly the same way as the tide on the sea-shore, the rise 

 and fall lasting from half an hour to an hour or more. This 



