August 6, 1903] 



NATURE 



335 



to a depth of 3(kk> metres. The copper sphere was sent 

 first to 3000 metres, but with no effect, and then to about 

 6000 metres, when the effect shown in Fig. 2 was produced. 

 The ratioiijle of the proceeding is : — at some depth less 

 than 3000 metres in the case of the brass tube, and less 

 than 6000 metres in the case of the copper sphere, the glass 

 tube in the former and the glass sphere in the latter case 

 collapsed suddenly. Considering, for brevity's sake, only 

 the brass tube ; immediately before the collapse the pressure 

 inside and outside the brass tube was equal and uniform. 

 The collapse of the glass tube produced a sudden and very 

 considerable relief of pressure inside the brass tube. 

 In ordinary circumstances the void so produced would 

 have been filled by water from the outside entering through 

 the perforated ends of the tube. But as the glass tube was 

 subjected to a pressure of nearly 300 atmospheres before it 

 collapsed, the difference of pressure produced in a moment 

 of time was between 200 and 300 atmospheres. The deep 

 corrugation shown in Fig. i proves that it was easier in the 

 time for the pressure to pinch up the stout brass tube than to 

 shove in the plugs of water at either end. The sudden 

 action of the pressure is due, not to the settling of the 

 column of 2000 to 3000 metres of water on the tube, but 

 to the resilience of the enormous quantity of water of high 

 tension produced by the pressure under which it finds itself. 



The effect produced on the copper sphere when the enclosed 

 glass sphere collapsed is of exactly the same kind. 



The experiment was originally made on board the 

 Challenger on the day after she made her deepest sounding 

 in the Atlantic in the neighbourhood of the West India 

 Islands: On that occasion both the thermometers attached 

 to the sounding line collapsed under the enormous pressure 

 of 3875 fathoms, amounting to 700 atmospheres, and 

 the experiment was made with tubes of three different 

 widths in water of 2800 fathoms in order to obtain data 

 for the construction of future thermometers. Two of the 

 tubes collapsed, only the narrowest, having a diameter of 

 6 millimetres, withstood both the pressure assisted by the 

 shock of the others collapsing near it. In all three cases 

 the glass tubes were converted into a fine powder like snow. 



The coiiapse of the brass tube, in the peculiar circum- 

 stances -of the experiment, is the exact counterpart of the 

 experiment which is frequently, but unintentionally, made 

 by people- out shooting, especially in winter. If, from in- 

 attention or. other cause, the muzzle of the gun gets stopped 

 with a plug of even the lightest snow, the gun, if fired 

 with this plug in its muzzle, invariably bursts. Light as 

 the plug of snow is, it requires a definite time for a finite 

 pressure, however great, to get it under way. During this 

 short time the tension of the powder gases becomes so 

 great that the barrel of the ordinary fowling-piece is unable 

 to withstand it and bursts. 



June 18. — " New Investigations into the Reduction 

 Phenomena of Animals and Plants." Preliminary Com- 

 munication. By Prof. J. B. Farmer, F.R.S., and J, E. S. 



In this communication the authors in the first place 

 pointed out that the attention which investigators have 

 recently paid to reduction phenomena occurring in animals 



Hg^.l. 



Pig. 2. 



and plants has resulted so far in an increasing divergence 

 of opinion, both respecting the nature of this process and 

 its significance. At the same time it was, however, 

 apparent that there were several important points upon 

 which all were now agreed ; it had, for example, been 

 clearly shown that, during this process, the number of the 

 chromosomes occurring in the cells affected was reduced by 

 one-half, and that this reduction was brought about during 

 the rest preceding two cell divisions, which appeared to 

 be invariably related to the process. Consequently it was 

 rendered probable that the explanation of reduction was 

 to be sought through a minute study of this, the synaptic 

 rest phase, in a number of selected animals and plants. 

 With this object, the authors had made a close examin- 

 ation of a large number of types, including mammals, 

 elasmobranchs, amphibia and insects among animals, 

 phanerogams, ferns and liverworts among plants, and the 

 results of this investigation are at variance with the 

 common existing conceptions of the process, while at the 

 same time they seem to indicate a possible reconcili- 

 ation between the different views which have been, and 

 still are, held by other investigators. It will be remembered 

 that there are two main theories of reduction. In the first 

 we have the process regarded as a qualitative division of 

 the chromatin by the separation into daughter nuclei of 

 entire somatic chromosomes. 



Fig:. 9. 



Fig. 4. 



NO. 1762, VOL. 68] 



In the second, the identity of the original somatic chromo- 

 somes becomes lost during the synaptic rest, and these are 

 then replaced by half the number of new ones, which, during 

 their formation, become longitudinally split twice in planes 

 at right angles to each other. This double longitudinal 

 division serves for two mitoses which take place almost 

 simultaneously. 



The authors find that at the end of the synaptic rest 



