^fAv, 1911. 



KNOWLEDGE. 



1S7 



as atropin or nicotin in Solanaceae ; or again they may be 

 chimaeras — sectorial, periclinal, or some other Icind. Carefnl 

 study shows that Solan u in tul)in}>ciisc, S. protetis, S, 

 KoelrcutcriaiiHDi. and S. Gacrtncrianitm are pericUnal 

 chimaeras. Of these forms, S. tiihiiigctise has a nightshade 

 bod}' and a tomato epidermis; S. Koclreiiteriaiiiim has the 

 reverse relation of the two components ; S. protcus has a 

 tomato periphery of two layers of cells, while S. Gacrtiieritm tini 

 apparently has the reverse relation. Winkler thinks, however, 

 that in S. Daruiiiiaiiiim he has a true graft hybrid produced 

 by fusion of the vegetative cells of the nightshade and tomato. 

 If this proves to be correct, it will be the first experimentally 

 produced graft hybrid, and the only instance of such a form 

 so far obtained. 



The wonderful discoveries of Winkler and Baur, which do 

 not appear as yet to have attracted attention in this country, 

 break new ground and open up a new field of experimental 

 biology. To both writers belongs equally the credit due to 

 the pioneer in a new branch of investigation. The bearing of 

 these discoveries upon theories of inheritance, and in particular 

 upon the function of the nucleus in heredity, brings us to the 

 cytological aspect of graft hybrids and chimaeras. In 1905 

 and 1907, Strasburger {Jahrb. fiir xciss. Bot., Band 42, Band 

 44) investigated Laburnum Adanii from this point of view. 

 If this form is really a hybrid — owing its origin to fusion of 

 the nuclei of Laburnum vulgare and Cytisus piirpurcus — 

 then its nuclei should contain a number of chromosomes 

 equalling the sum of the numbers characteristic of the 

 two parent species. This is not the case, however, and 

 Strasburger regarded this fact as evidence against the hybrid 

 character of the graft. He also investigated forms obtained 

 by grafting tomato and nightshade, and found that there was 

 no migration of nuclei and no fusion of the nuclei of stock and 

 scion ; hence he concluded that Winkler's graft hybrids are 

 merely chimaeras, calling them " hyperchimaeras " — forms in 

 which the elements of the two parent forms are more or less 

 intermingled but without any real nuclear fusion. Strasburger 

 therefore denies emphatically the reality of graft hybrids, but 

 as we have seen, Baur's results are of vastly greater import- 

 ance than these cytological observations, and they enable us to 

 place upon Winkler's work an interpretation different entirely 

 from those made by either Winkler himself or Strasburger. 



Strasburger goes on to consider the case of parasitism 

 between Angiosperms ; for instance, between mistletoe and its 

 host plants, where there is an intimate relation between the 

 two plants, but no mingling of nuclei. In grafting, a bud 

 from the point of union might possibly give rise to a shoot 

 bearing a flower in which an anther might be from the scion 

 and an ovary might be from the stock ; close fertilisation 

 might then give rise to a true hybrid, but, Strasburger argues, 

 hyperchimaeras would be more likely to produce flowers the 

 seeds of which would give rise to pure plants of either the 

 scion or the stock. The fact that pollen from his graft 

 hybrids would cause fertilisation in tomato or nightshade, 

 while neither of these plants can be crossed with the other, is 

 regarded by Winkler as proof of true hybrid character ; but 

 Strasburger thinks that the pollen was probably pure, 

 consequently fertilisation was to have been expected, but that 

 only nightshade or tomato would result. 



Winkler himself, in his account of the generation obtained 

 from the seed of his hybrids, gives some results as to the 

 chromosome numbers. In tomato the A' (sexual) and 2X 

 (somatic) numbers are 12 and 24, while in nightshade they are 

 36 and 72. He suggests that the difference in chromosome 

 numbers may prevent the crossing of the two species, though 

 noting the fact that Rosenberg crossed two species of Drosera 

 with 10 and 20 chromosomes in the sexual nuclei and obtained 

 a hybrid with 30 chromosomes as the 2A' number. If the 

 Solan uni hybrids are due to fusion of somatic nuclei, they 

 should have 72 + 24, or 96 chromosomes, unless the fusion 

 should be followed by reduction, in which case the number 

 would be 48. Winkler found the A' number to be 36 in 

 S. tubingcnse. S. Darwinianum. and S. Gaertncrianuni, 

 and found 12 in S. protcus and S. Koclrcutcrianum. the 



first three of these reverting in their pollen formation to night- 

 shade and the other two reverting to romato. Winkler suggests 

 that the graft hybrids more closely resembling nightshade are 

 from nightshade cells, and that those resembling tomato are 

 from cells of that parent, the nuclei being like those of one 

 parent or the other, but the protoplasm being mingled with 

 that of neighbouring cells. This theory, implying that the 

 protoplasm has great influence, obviously interferes with the 

 now generally accepted view that the nucleus is the sole 

 bearer of hereditary characters — but, as already stated, Baur's 

 results probably make both Winkler's and Strasburger's 

 explanations unnecessary. 



Much further work is required both on the experimental 

 production of graft hybrids and chimaeras, and the histological 

 and cytological characters of these forms. In his most recent 

 account of the cytology of the Solanum forms, Winkler (BtT. 

 dcutsch. bot. Gcs., 1910) states that S. tubingensc. S. 

 protcus. S. Koelrcuterianuni. and S. Gaertncrianuni are 

 periclinal hybrids; but that S. Darwinianum, at least in the 

 subepidermal layer of the stem apex, is a fusion hybrid. The 

 germ cells of S. Darwinianum have 48 chromosomes, and 

 since the parents (tomato and nightshade) have 12 and 36 

 chromosomes as the A numbers, \\'inkler infers that the 

 subepidermal layer from which the pollen is derived must 

 have 4S chromosomes ; he supposes that a nightshade cell 

 with 24 chromosomes has fused with a tomato cell with 72, 

 gi\ing a nucleus with 96. and that in the progeny of this 

 nucleus the number is reduced by halving. 



CHEMISTRY. 



By C. AiNSWORTH Mitchell, B.A. (Oxon.l, F.I.C. 



SPONTANEOUS COMBUSTION OF COAL. — The 

 various factors that tend to bring about the spontaneous 

 combustion of coal have been made the subject of an 

 experimental investigation by Messrs. Parr and Kressman, 

 who have published their results in the Journ. hid. Eng. 

 Chcm., 1911, III, 151. They show that oxidation processes 

 begin as soon as the coal is taken from the mine, and that 

 when the temperature produced by external factors reaches 

 a certain point, autoxidation sets in and results in the 

 ultimate destruction of the coal. The average temperature 

 for autoxidation lies between 200° and 275° C, according to 

 the state of division of the coal, while ignition takes place at 

 about 350' C. 



The factors contributing to raise the temperature of the 

 coal to the stage of autoxidation include : — (a) External 

 sources of heat, such as sunlight, or impact due to the 

 method of unloading; (b) Fineness of division; (c) Readily 

 oxidisable compounds of a bituminous nature; id) Iron pyrites, 

 the presence of as little of which as five per cent, may raise 

 the temperature by about 70' C. ; (e) Moisture, which 

 promotes the oxidation of the pyrites ; (/) Oxidation of carbon 

 and hydrogen, which takes place at temperatures above 120° to 

 140' C. ; (g) The fourth or autogenous stage of oxidation. 



Any measures to prevent spontaneous combustion must be 

 based upon a consideration of these facts. -All external 

 sources of heat must be eliminated as far as possible, and all 

 dust or finely divided material separated. Complete dryness 

 in storage will pre\-ent oxidation of iron pyrites, which is a 

 fruitful source of danger in coal from the mid-American fields. 

 No system of sorting at the mine can eliminate all risk from 

 this source. Drenching the coal with water may increase the 

 chances of oxidation where the sulphur is distributed through- 

 out the whole mass, while complete submersion of the coal 

 will probably not prove practicable. Preliminary heating 

 might be used to bring about the initial stages of oxidation, 

 and thus eliminate some of the factors which would 

 subsequently supply the necessary heat for destructive 

 oxidation processes. Treatment with chemical agents does 

 not hold out much chance of success, but a system of circulat- 

 ing a cooling li(]uid through pipes distributed throughout the 



