654 



NA TURE 



[NOVEMBI 



V-^3 



the re-attainment of the original siphon -length 

 depended on tl»e level at which the cut had been 

 made : it varic<l from it to 44 'lavs, with an average 

 of 27 dayt. Tl "1 upon once were 



kept under ob8< I 10 61 days after the 



original siphon - Imgili had Imjch re-attained, the 

 average period Iwing 42 days; those ooerated twice 

 for an average of 34 days ; and those tnree times for 

 27 days alter the last re-attainment of the original 

 siphon-length. In none of the operated animals did 

 any further growth of the siphons take place after 

 til' 1 dimensions had been reached. 



^ negative result of the preliminary experi- 

 mriii 11 siiiiied useless to try Dr. Kamnierer's further 

 operation of removing the gonads from the animals 

 with re-grown siphons, allowing other gonads to 

 regenerate, and then breeding a second generation. 



In 1013 it was shown at Naples that abnormally 

 long siplions of Ciona intestinalis can be grown by 

 keeping the animals in suspensions of abundant food 

 {Biol. Centrbl. 1914, vol. 34, p. 429). Were this the 

 reason for the long siphons of Dr. Hammerer's 

 operated Ciona, it should have been clear from 

 controls of unoperated animals kept in the same 

 water. 



A full account of the work at Roscoff will shortly 

 be published in the Journal of Genetics. 



H. MuNRO Fox. 



Zoological Department, Cambridge. 

 October 16. 



Selective Interruption of Molecular Movements. 



I WAS somewhat surprised to see that in spite of 

 Mr. Atkinson's letter, Mr. Fairbourne, in Nature of 

 July 21, still endeavours to maintain his view that the 

 relative gas-pressure in two communicating vessels 

 at equal temperature depends upon the shape of the 

 channel joining them, provided the pressure is suffi- 

 ciently low. The proper method of treating the 

 question, which of course does not lead to such an 

 extraordinary result, may be found in any text- 

 book on the kinetic theory, and it might have been 

 expected that Mr. Fairbourne, before claiming to 

 prove a paradox of this sort, would indicate in what 

 way the usual treatment is wrong. Instead of doing 

 so he adopts a curious treatment of his own, in which 

 he shows that in certain circumstances more paths 

 lead into one vessel than into the other, without con- 

 sidering that the number of molecules which enter 

 either vessel in unit time depends not only upon the 

 number of such paths but also upon their length. 

 When this is taken into account the usual result is 

 obtained, namely, that the pressure in the two vessels 

 is equal whatever the shape of the channel between 

 them. F. A. Lindemann. 



Clarendon Laboratory, Oxford, 

 October 15. 



Effects of Anaesthetics on Plants. 



Anaesthetics are known to cause alterations in 

 the permeability of cells to the ions of various salts. 

 It can be shown directly, by using the corolla of 

 ipomoea Learii, that the permeability of plant cells 

 to carbon dioxide is also altered by anaesthetics. 



The corolla consists of two layers of cells only, with 

 thin cuticle, no intercellular spaces, no stomata. The 

 cell-sap contains an anthocyanin, which indicates 

 Pi, (thus avoiding complications due to an added 

 indicator). The buds are pink (Pn 6), changing to 

 full blue (Ph 7'8) as the flower opens, in 30-40 

 minutes. The blue changes through violet to pink 

 as the flower withers (6-8 hours). Portions of the 

 blue corolla floated on water saturated with carbon 



NO. 2818, VOL. 1 1 2] 



dioxide rapidly turn pink : this change is reversible 

 on removing to plain water. 



If discs cut from the corolla are first treated with 

 aoueous solutions (o'04M-o'iM) of chloroform or 

 etner. and then with a .saturated solution of carbon 

 dioxide (Pm 5), a time-curve can be constructed, 

 showing the changes in permeability to carbon 

 dioxide induced by varying exposures to the anss- 

 thetics. 



The first effect is a marked decreas* ' ich 



as 50 per cent.) in the rate of peneti m 



dioxide into the cell, the decrea.s< 1(^-15 



minutes ; the rate then increases r. .iching 



200 per cent, in 40 minutes, and oootinumK <.'. lacrease. 

 After 40-50 minutes' exposure to the anaesthetics the 

 tissue Decomes irresponsive. 



In order to reach the cell-sap the carlmn dioxide 

 must pass through (1) the cell-wall, (2) the protoplasm 

 lining the cell-wall. The fact that the cells of the 

 disc change colour simultaneously shows that the 

 carbon dioxide passes freely through the wall. On 

 the other hand, hydrochloric, sulphuric, and acetic 

 acids of the same Pn as the carbonic (P„ 5) penetrate 

 only from the cut edges of the discs inwards, and not 

 over the whole area. The addition of ether or chloro- 

 form to these acids has a similar effect on their rate 

 of penetration into the cell as on carbonic acid. It 

 is therefore concluded that ether and chloroform alter 

 the permeability of the plant cell to carbon dioxide by 

 their action on the protoplasm and not on the cell-wall. 



These alterations in permeability to carbon dioxide 

 may affect the apparent rate of respiration (measured 

 as carbon dioxide output) under anaesthetics, and 

 a suitable correction may require to be made in such 

 experiments. E. Philip Smith. 



46 Murrayfield Avenue, Edinburgh, 

 October 13. 



Stereoisomerism among Derivatives of Diphenyi. 



The references to Dewar's formula for benzene 

 which are made in the letters of Dr. Turner and Dr. 

 Kenner in Nature of September 22 and October 13 

 (pp. 439 and 539) raise a point of some importance 

 in regard to the use of symbols in chemistr>\ Sir 

 William Bragg 's work has revealed the fact that the 

 length of the carbon-to-carbon bond is remarkably 

 constant at about 1*5 A.U. Dr. Turner, however, 

 following the common convention, represents the 

 para-linkage in his formula for diphenyi by a bond 

 which is perhaps twice as long as those joining 

 adjacent atoms in the ring. It is, of course, possible 

 to maintain the normal length of the bond by dis- 

 torting the benzene hexagon into a quadrilateral, thus, 



H 

 HC— C— CH 



H 



H 



H 



but there is, I believe, no indication whatever of any 

 such extreme distortion in "Bragg 's work on the 

 X-ray analysis of crystals of aromatic compounds. 

 This difficulty would not arise if the para-linkage were 

 regarded as indicating only the existence of " free 

 affinities" on the i and 4 carbon atoms, or of an 

 electro valency between them ; but so long as this 

 link is treated as a real bond, there does not seem to 

 be any justification for stretching it to an abnormal 

 length, although this is clearly necessary in order to 

 presers'-e the very well founded idea that the benzene- 

 ring is fundamentally hexagonal in form. 



T. M. LowRY. 

 The University, Cambridge, 

 October 18. 



