494 



NATURE 



[December 30, 1915 



numbers on a single host. The development of these 

 has hitherto been something of a puzzle, since all the 

 individuals found on one host are approximately in the 

 same stage of development, and simultaneous infection 

 by a swarm of larvae seems improbable on account of 

 the rare occurrence of the parasites. Mr. Geoffrey 

 Smith, in 1906, made the suggestion that these 

 gregarious parasites might be produced by some pro- 

 cess of polyembryony or budding in the endoparasitic 

 stage. He' based the suggestion on the fact, observed 

 by Delage and confirmed by himself, that the develop- 

 ing Sacculina may, as a rare exception, produce twin 

 rudiments of the body within a single embryonic mass. 

 During the recent expedition sent by the Carnegie 

 Institution of Washington to Torres Straits, Mr. F. A. 

 Potts, of Cambridge, was able to investigate some of 

 these " social " species of Rhizocephala and to obtain 

 striking and conclusive evidence of the truth of this 

 suggestion. The forms which he studied belong to 

 the genus Thompsonia of Kossmann (with which he 

 identifies Thylacoplethus of Couti^re), and are para- 

 sites on various tropical crabs and shrimps. The 

 saccular bodies occur in hundreds on the surface of the 

 body and limbs of a single host, and are of much 

 simpler structure than in the case of Sacculina. There 

 is no mantle-cavity, nor are there genital ducts or 

 associated glands, and the nervous system, reduced to 

 a vestige in Sacculina, has entirely disappeared. 

 Further, while Sacculina, like most cirripedes, is her- 

 maphrodite, Thompsonia has no trace of male organs, 

 and there is reason to believe that it reproduces solely 

 by parthenogenesis. The sac-like bodies contain 

 nothing but a mass of developing eggs or larvae, which 

 are set free in the so-called " cypris " stage, the earlier 

 nauplius stage being suppressed. Mr. Potts was able 

 to show that all the external sacs are in connection 

 with a single continuous system of roots, and, further, 

 that when the sacs are cast off by the moulting of the 

 host (probably after their contained larvae have been 

 matured and set free), a fresh crop of sacs is budded 

 off from the root-system. 



It would scarcely seem possible to imagine a wider 

 departure from our ordinary conception of an arthropod 

 than that presented by Thompsonia, with its " myce- 

 lium-like root-system producing its singular asexual 

 reproductive organs." Mr. Potts directs attention to 

 the analogies it shows with certain Hvdromedusae. As 

 if to complete the resemblance, he finds evidence that 

 germ-cells are in some cases produced within the root- 

 system, although he was unable to discover whether 

 they migrate from thence into the developing buds as 

 they migrate from the coenosarc into the sexual buds 

 of some Hydromedusae. W. T. C. 



CONDUCTION OF ELECTRICITY 



THROUGH METALS.^ 



'T'HE power of transmitting large electric currents 



-* is one of the most characteristic properties of 

 metals, and one to which they owe no insignificant 

 fraction of their industrial importance. If we imagine 

 for a moment the revolution that would be made in 

 our daily life if metals did not possess this property, 

 how much we rely upon it for light, locomotion, and 

 communication, we shall realise how large a part it 

 plays in our social life. It is not, however, on this 

 aspect of metallic conduction that I wish to speak 

 this evening, but rather on the mechanism by which 

 the flow of current is produced, and the light which 

 electric conduction throws on the structure of metals. 



It is remarkable that though the quantity of elec- 

 tric current which flows any day or week through 



1 Lecture delivered before the Institute of Metals by Sir J. J. Thomson, 

 O.M., F.R.S. y . 'J J 



NO. 2409, VOL. 96] 



liquids or gases is quite insignificant in comparison 

 with that which flows through metals, our views of 

 the nature of conduction through gases or liquids 

 are much more settled and definite than any we 

 possess with respect to metals ; it is remarkable, too, 

 that progress in this subject is the outcome of the 

 study of the flow of electricity through gases rather 

 than through liquids, and that it is gases and not 

 liquids which have given us the clue which promises 

 to lead to the solution of the conduction of electricity 

 through metals, which from many points of view is 

 by far the most important case of conduction. 



Many physicists have had the idea that the passage 

 of electricity through metals might be analogous to 

 that through liquids, where we have strong evidence 

 that the current is carried by atoms or groups of 

 atoms charged with electricity. Some of these atoms 

 are charged with positive electricity, others with nega- 

 tive, but all help to carry the current, the positively 

 charged ones by moving in the direction of the cur- 

 rent, the negatively charged ones in the opposite 

 direction ; in the case of liquids the passage of the 

 current is inextricably connected with the movement 

 of atoms through the liquid. A familiar instance of 

 this is 'the ordinary electrolytic bath for electro- 

 plating. Now, as I have said, some physicists had 

 the idea that electric currents get through metals by 

 the motion of charged atoms in much . the same way 

 as they get through liquid, and they naturally made 

 experiments to see if they could detect any transport 

 of metal which is so marked a feature when currents 

 pass through liquids. Speaking generally, these 

 experiments were of two types. In one type the 

 current was sent through a wire made of an alloy 

 of two metals. The proportion of the two metals in 

 the wire before a current was passed through it was 

 determined with great care. A powerful current was 

 then sent through the wire for a long time, and the 

 wire was again analysed, samples being cut from 

 the end at which the current went in and also from 

 the end at which it went out. If there had been any 

 transport of atoms by the current, the composition 

 of the end where the current entered would be 

 different from that of the end where it left. The 

 experiment showed that no difference could be de- 

 tected, even when enough current had passed through 

 the wire to carry all the metal in the wire from one 

 electrode to the other if the metal had been in a salt 

 dissolved in an electrolytic bath. The other type of 

 experiment was to put plates of two metals, say 

 gold and lead, together, taking care that there was 

 good contact, and then send a strong current across 

 the junction for a considerable time, and at the end 

 of that time see if any lead had been carried into 

 the gold or any gold into the lead. Not the slightest 

 trace of any such transport could be detected. These 

 experiments showed that the electricitv was not 

 carried through metals by charges on atoms or any 

 combination of atoms, and as at that time no other 

 carriers of electricity were known, the difTficulties 

 in the way of supposing that the current was carried 

 by moving electric charges seemed insuperable. 

 Relief, however, came from the study of the passage 

 of electricity through gases, which showed that there 

 were other carriers of electricity besides atoms. 



These carriers, which are called corpuscles or elec- 

 trons, are exceedingly small compared with the smallest 

 atom known, that of hydrogen. They form a part 

 of every kind of atom, and, however different the 

 atoms from which the electrons come may be, the 

 electrons themselves are invariable. There is only 

 one kind of electron, and this has a mass of ^700 

 that of a hydrogen atom, and carries a , constant 

 charge of negative electricity. The discovery of the 

 corpuscle or electron put an entirely different com- 



