124 



NATURE 



[April 



19 '5 



discusses the geographical distribution of the west 

 German Unionida;, including extinct forms. 



The manner in which spiders make their webs forms 

 ubjed ol editorial notes in the December number 

 of The Country-Side. It is stated that all the pub- 

 lished accounts which have come under notice describe 

 spiders as constructing their webs in narrowing circles 

 from the periphery towards the centre; but first-hand 

 of the opposite mode of procedure (thai is, 

 working from the centre outwards) is cited, and the 

 concludes that the published descriptions refer 

 only to the repair of broken webs. 



In the introduction to an elaborate monograph of 

 the crinoids of the Indian Ocean, forming part 7 of 

 " Echinoderma of the Indian Museum," published at 

 Calcutta, Mr. A. H. Clark dwells on the extreme 

 richness of the crinoid fauna of this area, which he 

 regards as representing the stock that has given 

 origin in the past to similar faunas in many other 

 parts of the world. Nearly 400 Indian forms are now 

 known, of which about 750 are comatulids and the 

 remainder stalked types. They are arranged in nineteeen 

 families, with eighty-two genera, all the species being 

 peculiar to the Indian region. The only family absent 

 from this is the monogeneric Holopidas. "All the 

 genera of the Atlantic, Antarctic, and Arctic Oceans 

 arc closely related to East Indian genera, from which 

 they were evidently derived in the remote past; but 

 in many cases a single East Indian genus has 

 apparently given rise to two or more Atlantic genera, 

 all nearly equally related to the parent stock." It is 

 also_ stated that crinoids may be utilised for obtaining 

 an idea of the nature of the plankton of the seas in 

 which they grow, thus affording a clue as to the 

 suitability, or otherwise, of any given area for the sup- 

 port of food-fishes, sponges, coral, or pearl-oysters. 



The fifth part of vol. x. of the Annals of the South 

 African Museum is devoted to an account by the "Rev. 

 T. R. R. Stebbing of the local representatives of the 

 group of small marine crustaceans known as 

 Svmpoda, or— if we follow the Cambridge Natural 

 History— Cumacea. Although the members of the 

 group are readily distinguishable from other crusta- 

 ceans, their classification is a matter of difficulty, 

 owing to the interlacing of characters and the exist- 

 ence of fine gradations. The author, who recognises 

 a larger number of families than is adopted in the 

 work cited, describes nine genera and fourteen species 

 as new. 



The pseudo-scorpions of the country form the sub- 

 ject of vol. x., part 4, of the Annals of the South 

 African Museum. According to the author, the Rev. 

 E. Ellingsen, less than half-a-dozen local representa- 

 tives of the group were known at the beginning- of the 

 century, but the list is now very large, and has been 

 increased in the article before us. The type genus, 

 Chelifer, it is pointed out, will ere long have to be 



R. L. 



divided. 



FOAM STRUCTURE OF METALS. 



J N a paper on the "foam structure" of metals, in 

 The International Journal of Metallography liii., 

 1), Prof. Quincke gives a summary of the conclusions 

 winch he states as the result of researches dating 

 from 1S58 to the present day. While Prof. Ouincke's 

 views may well claim respectful consideration, his 

 statement of them in the present paper is far 'from 

 convincing, and his effort to extend to metals his 

 theory of foam structure of matter appears to be 

 singularly strained. To begin with, there is the 

 fundamental assumption that before solidification 

 commences even in a "pure" metal the liquid be- 

 NO. 2266, VOL. qi] 



comes heterogeneous, being divided into foam-cells by 

 minute cell-walls differing in viscosity and surface- 

 tension from the cell-contents. Quincke supposes 

 these to be so minute that experimental evidence of their 

 existence cannot be obtained, and he depends for the 

 justification of his assumption upon the power of his 

 theory to explain all the known phenomena of the 

 structure and properties of metals. The present paper 

 gives an outline of this explanation, but while it is 

 distinctly ingenious it suffers from the defect that its 

 author is obviously incompletely acquainted with the 

 modern developments of metallography. As a result, 

 one finds again and again that the proffered explana* 

 tions are incompatible with well-established facts. 

 One example, out of many which might be given, 

 must suffice. 



According to Quincke, the growth of crystals during 

 annealing is due to the collapse of a foam-wall lying 

 between two adjacent foam-cells, and forming what 

 is usually termed an intercrystalline boundary, with 

 the consequent coalescence of the two adjacent crystals 

 into a single crystal. Direct observation of the pro- 

 cess of crystal growth has, however, definitely shown 

 that this is not the true modus operandi. The crystals 

 do not grow bv the bodily absorption of their neigh- 

 bours, but by a process which may be likened to 

 gradual invasion and conversion. The growing 

 crystal gradually pushes its boundary outward into 

 its neighbours, and frequently does so by pushing 

 out one or more arms which gradually spread laterally 

 as well as advance longitudinally. Nothing could be 

 more unlike the picture suggested by Quincke's ex- 

 planation, and similar difficulties can be raised at 

 every turn. 



On reading the paper, however, while those inti- 

 mately acquainted with the behaviour of crystalline 

 aggregates will scarcely be disposed to accept the 

 " foam-cell " theory, they will yet be struck by the 

 fact that the forces of surface-tension upon which 

 Quincke lays such stress must powerfully affect the 

 structure of metals and alloys — forces the importance 

 of which has not perhaps been sufficiently recognised 

 by current metallographic theories. In eutectic alloys 

 particularly one constantly meets with structures 

 which bear strikingly close resemblance to those 

 assumed by films of liquid under the action of surface- 

 tension. It has even been thought that the con- 

 stituents of such eutectics may assume their actual 

 forms just before solidification, in the shape of bags 

 1 or sacks of the kind imagined by Quincke as foam- 

 j cells. 



Experimental evidence is, however, against this 

 view. The experiment has been tried of allowing 

 eutectic alloys to solidify slowly under the action of 

 centrifugal pressure in a powerful centrifuge, and 

 the resulting structure is entirely unaffected. Had 

 liquid sacks or " foam-cells " really been formed thev 

 must have been flattened or deformed under this 

 treatment, but such was not the case. On the other 

 hand, recent metallographic researches seem to indi- 

 cate that the intercrystalline boundaries of a metal 

 are of the nature of cell-walls formed by very thin 

 layers of the same metal in the amorphous or under- 

 cooled liquid state, and here there is a decided approxi- 

 mation to Quincke's ideas, onlv that these cell-walls 

 are regarded as the result of the meeting of adjacent 

 growing crystals, and not as the primary limitations 

 to crystal growth. Still, although Quincke's theory 

 of foam-cells can scarcely be accepted as being in 

 reasonable accordance with the known facts of 

 metallography, a study of his views should be useful 

 and suggestive to all those interested in the physics 

 and physical chemistry of crystalline aggregates. 



W. ROSENHAIN. 





