ii8 



NATURE 



[May 29, 1890 



anatomical characters of the Hamamelidacese, examined with the 

 object of using them as a basis for the systematic arrangement 

 of the family, by A. Reinsch (i plate). — A list of the Poly- 

 podiacese. Graminese, Cyperacere, and Juncacea?, collected by 

 Dr. Marloth in South Africa. — On Cissine domingemis Spr., by 

 A. Garcke. — A treatise on the genus Platanus, with two plates, 

 hy J. Janko. This paper has special reference to the detail 

 characters of the leaf. — On two Soldanellas new to the flora of 

 Hungary, by V. A. Richter. — At the end of the volume are 

 abstracts of many recent memoirs published elsewhere, and 

 a classified list of the most important works on systematic, 

 geographical, and descriptive botany published in the year 

 1889. 



Bulletin de la Societe des Naturalist es de Moscou, 1889, No. 2. 

 — On the origin of periodical comets, by Th. Bredichin (in 

 French). The author examines into the cases of division of 

 comets into two or more individuals, endeavouring to classify the 

 better known ones into "families," and gives the formulae for the 

 cases when the impetus given to the corpuscles of a comet acted 

 under a given angle to the plane of its motion. — Note on the 

 genus Bovibtis, by General Radoszkowski (in French). — The 

 Amphibian fauna of Europe : \}a&Anura, by Dr. J. Hedriaga, being 

 a full description (in German) of the two genera Kana and Bufo, 

 their species, varieties, life, and geographical distribution. 



No. 3. — On the modes of propagation of fresh-water fishes, by 

 S. Nikitin (in French). M. Nikitin objects to the too hasty gene- 

 ralizations sometimes arrived at by men of science (especially with 

 regard to Central Asia) as to the former communication between 

 distant fluviatile basins and lakes which now have some species 

 of fishes in common. He points out the possibility of the 

 transport of the eggs of certain fishes by birds, and mentions 

 the fact of young pikes, from six to ten centimetres long, being 

 found in small temporary ponds on the banks of the Moskva 

 river, where they could by no means have migrated themselves. 

 Further inquiry is asked for. — The transport of electrical energy, 

 by J. Weinberg (in German). — On the nesting of Podoces 

 Panderi, by N. Zarudny. — The Amphibian fauna of Europe : 

 the Amira (continued), by Dr. J. Bedriaga. The genera Hyla, 

 Pelobatiis, Pelodytes, Dlscoglossus, Bombinatur, and Alytes, are 

 considered, and the author describes two new varieties of Hyla 

 a)-borea under the names of var. orientalis and var. Molleri. — 

 On the influence of weather upon plants and animals, by Alex. 

 Becker (in German). 



SOCIETIES AND ACADEMIES. 

 London. 



Royal Society, May 8. — " Experiments on Vapour- 

 Density." By E. P. Perman, B.Sc. Communicated by Prof. 

 Ramsay, F.R.S. 



Vapour -density of Bromine. — This work was undertaken in 

 order to see if the results of Prof. J. J. Thomson's experiments 

 could be verified ; these appeared to show that bromine vapour 

 dissociated on continued heating at a low pressure, and a compara- 

 tively low temperature. The method used was a modification of 

 the Dumas method, by which a series of vapour-density determina- 

 tions were made at different pressures, with the same identical 

 material. The chief conclusions arrived at are (i) that no 

 dissociation takes place at temperatures as high as 280°, and 

 pressures as low as 20 mm., even on continued heating ; (2) that 

 bromine vapour has no tendency to form molecules with more 

 than two atoms, on approaching the liquid state. 



Vapour-density of Iodine. — The density of saturated iodine 

 vapour was determined by an adaptation of Kundt's method of 

 determining the velocity of sound in gases. The mean result 

 was 126 "g, showing that liquid iodine has the formula Ij. 



Induction Spark through Iodine Vapour. — The same apparatus 

 was used as in finding the vapour-density of iodine. No altera- 

 tion of the wave-length in the iodine vapour (as indicated by the 

 heaps of finely-divided silica on the lower part of the tube) 

 ■occurred on passing a series of sparks, and then causing the glass 

 piston to vibrate. Sparking does not appear, therefore, to 

 produce permanent dissociation of iodine vapour, notwithstand- 

 ing the results pointing to a contrary conclusion, obtained by 

 Prof. Thomson. Neither bromine nor iodine vapour, when 

 saturated, threw the silica into heaps ; it appears that sound- 

 waves cannot be propagated in a saturated vapour, for con- 

 densation will be produced either by the waves of compression 

 ■or those of expansion (according to the nature of the vapour, 



NO. 1074, VOL. 42] 



and its temperature) and the rates of propagation of the two sets 

 of waves will therefore be different. 



Vapour-densities of Sulphuric Anhydride, and Aqueous Hydro- 

 chloric Acid. — These were determined by the same method as 

 the vapour-density of bromine. The vapour-density of sulphuric 

 anhydride indicated a formula SO3, and that of a<iueous hydro- 

 chloric acid showed that it is a niixture of molecules HCl and 

 HjO, and not a compound. In all these experiments the 

 quantity of substance in the globe was not found by weighing, 

 but by estimating it volumetrically, portions being drawn off and 

 absorbed in a suitable liquid. The globe was heated by means 

 of a vapour-jacket ; the vapours used were those of alcohol, 

 chlorobenzene, bromobenzene, and bromonaphthalene. 



The author is greatly indebted to Prof. Ramsay for constant 

 advice and assistance in carrying out the work. 



May 22. — " The Chemical Products of the Growth of Bacillus 

 anthracis and their Physiological Action." Hy Sidney Martin, 

 M.D., Pathologist to the Middlesex Hospital. Communicated 

 by Dr. Klein, F.R.S. 



The bacilli were grown in a solution of pure alkali-albumin 

 (made from serum-proteids) and of mineral salts of the com- 

 position of the salts of the serum. 



The cultivation of the bacilli was continued for ten to fifteen 

 days, and the organisms removed by filtering through Chamber- 

 lain's filter. The filtrate contained the products of the bacterial 

 growth, viz. : — 



(i) Proto-albuTnose and deutero-albumose, and a trace of 

 peptone: all with the same chemical reactions as the similar 

 bodies formed in peptic digestion. 

 {2) An alkaloid. 



(3) Small quantities of leucin and tyrosin. 

 The chief characteristic of the anthrax proto- and deutero- 

 albumose is their strong alkalinity in solution —an alkalinity not 

 removed by absolute alcohol, by benzene, chloroform, or ether, 

 or by prolonged dialysis. Acid-alcohol dissolves from the 

 alkaline albumoses a trace of a poisonous body, but this is not 

 in proportion to the toxicity of the albumoses. The albumoses 

 are precipitated in an alkaline condition by saturation with 

 NaCl (proto-albumose) or (NH4).^S04. The alkaloid is soluble 

 in absolute alcohol, amyl alcohol, and in water ; insoluble in 

 benzene, chloroform, and ether. It is strongly alkaline in solu- 

 tion, and a powerful base, readily forming salts with acids. The 

 sulphate crystallizes in small needles or prisms ; the oxalate 

 in long, branching needles or flat plates. From the salts the 

 alkaloid is easily regained. In solution, the alkaloid is precipi- 

 tated by phosphotungstic, phosphomolybdic, and phosphoanti- 

 monic acids and platinic chloride, but not by potassio-mercuric 

 iodide. It is slightly volatile, and, when kept exposed to the 

 air, it becomes acid, and loses, to a great extent, its poisonous 

 properties. 



Physiological Action, 



(i) The mixture of anthrax proto- and deutero-albumose is 

 poisonous. In small doses it produces in mice a local subcuta- 

 neous oedema, with some sluggishness, ending in recovery. 

 Larger doses produce a greater oedema with more signs of ill- 

 ness, sluggishness leading to prolonged stupor, coma, and death 

 in twenty-four hours or longer. A fatal dose for a mouse of 22 

 grams weight is o'3 gram (subcutaneously injected). In some 

 cases the spleen is enlarged : no organisms being present, as 

 shown by gelatine tube cultivations. Boiling for a short time 

 diminishes the toxicity of the proteid, but does not completely 

 destroy it, and death may result from the boiled albumoses. 



(2) The anthrax alkaloid produces symptoms and lesions simi- 

 lar to the albumoses, but much more rapidly and severely. The 

 animal becomes ill directly after the injection, gradually becomes 

 more and more sluggish, and dies in coma, or, if a non-lethal 

 dose be given, it recovers from the state of stupor gradually. 

 After de:\th, enormous local subcutaneous oedema is found, with 

 congestion and sometimes thrombosis of the small veins. Peri- 

 toneal effusion is occasionally present, and the spleen is usually 

 enlarged, dark, and congested, or simply congested without 

 being greatly enlarged. The fatal dose for a mouse weighing 

 22 grams is between o-i and 0"IS gram, death occurring in two 

 to three hours 



The anthrax bacillus in digesting the alkali-albumin forms (i) 

 proto-albumose, (2) deutero-albumose, (3) an alkaloid. The 

 alkalinity of the albumoses may explain their toxic properties, 

 being due to the fact that the alkaloid is in a "nascent " condi- 

 tion in the albumose molecule. The bacillus forms the alkaloid 



