May 13, 1922] 



NA TURE 



613 



which was finally lost sight of ; but a complete 

 division could not be stated to have been seen in this 

 case, and other observations point to the formation of 

 long protoplasmic connexions between the leucocytes of 

 the oyster as a.repair phenomenon somewhat equivalent 

 to the clotting effect of blood in vertebrates. The 

 cultivation of the leucocytes of the oyster may never- 

 theless be possible, since Carrel and Burrows and later 

 workers have shown that even specialised tissues of 

 vertebrates can be grown outside the body in media 

 and under conditions now well-known. A successful 

 cultivation of the leucocytes of the oyster would 

 undoubtedly suggest methods of attacking the 

 problem of the cultivation of human leucocytes out- 

 side the body under known conditions and to yield 

 known properties in the leucocytes. The potenti- 

 alities of human leucocytes cultivated under such 

 conditions may be so great that sufficient excuse is 

 provided for any speculation which may point even 

 faintly to a method by which such a desirable product 

 may be attained. J. H. Orton. 



The Marine Biological Laboratory, 

 The Hoe, Plymouth, April 28. 



Periodical Phenomena in the Temperature Functions 

 of Certain Properties of the Metals. 



It is well known that the assumption of a quantum 

 distribution of the energy of vibrations of the atoms 

 in solids can be used to explain changes of the 

 specific heat and other properties at low tempera- 

 tures. Now this quantum distribution also seems 

 to reveal itself in a more direct manner. 



I have measured, partly in collaboration with Mr. F. 

 Gunneson, the thermoelectric force at the ordinary 

 temperature of specimens of iron and tungsten that 

 had been successively heated at different temperatures 

 and each time rapidly cooled. The thermoelectric 

 force taken as a function of the heating temperature 

 shows periodical changes, and I could easily distin- 

 guish certain repeated intervals of transformation. 

 The approximative absolute temperature, T„ of 

 transformation, satisfies the relation T„ =A . «, 

 where A is a constant and n an integer number. 

 This relation has been verified experimentally for 

 iron from w =3 to 12, and for tungsten from w =4 

 to 9. The mean value found for A was for pure 

 iron 97° and for tungsten 82°. 



As remarked above, this phenomenon is most 

 probably due to the quantum distribution of vibra- 

 tion energy. As the mean number of quanta of the 

 atoms, however, increases continually with the 

 temperature, I have assumed, to explain the period- 

 icities, that the transformations occur at every 

 temperature for which a certain fractional part 

 (i/y) of the atoms has a number of quanta like, or 

 greater than, a new integer. This assumption gives 

 A=6»/3logy, where e[^^v) is the characteristic 

 temperature of the metal in consideration. From 

 this relation r may approximately be calculated, 

 though the values are, like those of 0, rather un- 

 certain. I found for iron r—j^-b, and for tungsten 

 y=3-3. Thus the critical fractional part is about 

 one-fourth, which seems to be a reasonable value. 



G. BORELIUS. 



Physical Institute, University, Lund. 



Observation of Comets. 



Dr. Crommelin in his valuable " Comet Notes," 

 B.A.A. Journal, March 1922, p. 198, speaking of 

 Reid's comet (1922 a), says " the comet must have 

 been quite well placed for Northern observers last 

 October and November, being of the 9th magnitude." 

 It seems that Mr. Reid discovered the comet only a 



NO. 2741, VOL. 109] 



short time before it passed out of sight. This in- 

 cident reminds us of the fact that a considerable 

 number of these objects must escape observation 

 altogether owing to the want of observers. It is 

 remarkable that English astronomers appear hitherto 

 to have taken little interest in cometary work, and 

 that very few comets have been discovered from this 

 country. 



This is a regrettable circumstance. Yet the dis- 

 covery and observation of comets include a com- 

 paratively easy and very attractive field of work open 

 to amateurs with moderately powerful instruments. 

 There are a. great number of telescopic observers in 

 the United Kingdom who have the means and the 

 time at their disposal to accomplish valuable work 

 in this department if they would only engage in it 

 in an earnest manner. It is a branch which offers 

 special inducements to amateurs, and holds out a 

 greater prospect of brilliant success than perhaps any 

 other sphere of labour. It is hoped, therefore, that 

 some enthusiasts will turn their practical attention to 

 it, for it is fair to suppose that some of them would 

 Uke to follow in the footsteps of Messier and Pons, 

 and the equally famous modem discoverers of comets, 

 Barnard and Brooks. 



The. cometary section of the B.A.A. could not have 

 a more capable director than Dr. Crommelin, and it 

 would strengthen his hands, provide material for his 

 researches, and repay him for his labours if the 

 members of the section notified him of their discovery 

 of one or two comets every year. 



W. F. Denning. 



44 Egerton Road, Bishopston, Bristol. 



A Proposed Laboratory Test of the 

 Theory of Relativity. 



I SHOULD like to comment on the assumption as to 

 the atomic weight of RaG made by Dr. King in his 

 letter in Nature of May 6, p. 582. Instead of 

 taking this as known, and calculating the atomic 

 weight _of radium by taking account of the a-particle, 

 energy and electron masses, it appears to me that the 

 reverse process would be more justified. It is true 

 that there is an unexplained discrepancy in the 

 difference between the atomic weights of uranium 

 and radium, but we have no reason to doubt the 

 value of the atomic weight of radium obtained by 

 Honigschmid. As his radium was obtained from 

 Joachimstal pitchblende, which contains practically 

 no thorium, there is no liability to disturbance from 

 the presence of isotopes of radium in his material. 

 On the other hand, the theoretical atomic weight of 

 RaG was obtained by subtracting the mass of five 

 a-particles from the atomic weight of Ra. That the 

 lowest value (206-046) found by Honigschmid for 

 lead from Morogoro pitchblende agrees so nearly 

 with the theoretical value (205-93), shows only that 

 the amount of original lead present in this material 

 was very small, although, of course, it gave splendid 

 confirmation to the theory of isotopy. We cannot 

 say, however, that we are dealing with pure RaG, 

 even though we know that the amount of admixed 

 isotopes must be very small. 



In dealing with such small corrections to the atomic 

 weight as are involved in Dr. King's letter, it thus 

 appears unjustifiable to start off by assuming that 

 the atomic weight of RaG is 206-00. As Dr. King 

 suggests, however, when the technique of the deter- 

 mination of mass spectra has developed, we shall be 

 independent of admixed isotopes, and the problem 

 will then be put to a rigorous test. 



Robert W. Lav^tscn. 



The University, Sheffield. 



X 2 



