December 12, 1907J 



NA TURE 



127 



to apply to mulattos, and some doubt is thrown on the 

 case cited by Wells. On the other hand, Lawrence quotes 

 from the Philosophical Transactions (" v., 55 ") a case of 

 two negroes who had a white child, the paternal grand- 

 father being white. This seems purely Mendelian. 



November 25. W. T. Thiselton-Dver. 



Specific Stability and MuTation. 



The desire to be as brief as possible has led, I fear, to 

 some obscurity in the sentences quoted by Sir William 

 Thiselton-Dycr (p. 77) from my letter of October 17. The 

 meaning will perhaps be clearer if I explain the precise 

 significance which 1 attached to the words " appear " and 

 " occurrence." 



By the occurrence of a mutation in one of the higher 

 plants I meant the production of a seed capable of 

 germination and containing an embryo with definitely 

 different potentialities from those of its parent. The 

 appearance of a mutation, on the other hand, implies that 

 such a seed has germinated and given rise to a plant 

 recognisably different from other members of the species. 

 My contention is that the conditions of cultivation are such 

 as to allow of the safe germination and growth of plants 

 which would have no chance of survival under natural con- 

 ditions. It is therefore possible that mutations may occur 

 as frequently under natural conditions as under cultivation. 

 This being so, it does not appear to me to be an abuse 

 of language to state that the assumption that cultivation 

 causes the occurrence of mutations is one which requires 

 proof. In support of this assumption Sir William 

 Thiselton-Dyer brings forward certain evidence. With 

 much of this evidence I was already familiar, but it did 

 not appear to me to amount to satisfactory proof of the 

 current position. The authority of eminent breeders is 

 quoted for the fact that, as soon as one new variety of a 

 cultivated species has been obtained, a host of others 

 immediately follow. But the explanation of this may be 

 that the breeder, as soon as he has obtained a single 

 novelty, immediately crosses it — deliberately or by accident 

 — with the original type, thus giving rise to endless new 

 combinations. R. H. Lock. 



Botany School, Cambridge, December 2. 



TiHT mutations inevitably appear sooner or later under 

 cultural conditions is not an assumption, but a fact. 

 That they do so only casually under natural conditions, and 

 usually fail to perpetuate themselves, equally seems to me 

 not an assumption, but a fact. If, as Mr. Lock seems to 

 argue, there is an equal chance of their occurence in 

 either case, then their appearance should be more frequent 

 in nature than in cultivation, because the former has a 

 larger population to work with. But it is not so. I 

 therefore conclude with Darwin that cultivation introduces 

 some provocative condition which is lacking (or latent) in 

 nature. What that condition is seems to me a very 

 important subject for research. 



December 5. W. T. Thiselton-Dver. 



The Winding of Rivers. 



With your permission I would like to make a few re- 

 marks on the winding of rivers, which is at present being 

 discussed in your pages. My observations were made while 

 fishing, and my remarks refer to the rivers of our own 

 country, and may not apply to rivers of greater volume. 

 But first I would like to point out an objection to Prof. 

 J. Thomson's experiments. In Prof. Thomson's paper in 

 the report of the British .■\ssociation for 1876 no details of 

 the conditions of the experiment are given, but Sir Oliver 

 Lodge in his letter (N.ature, November 28) says Prof. 

 Thomson's model had a wooden bed. Now it is 

 very evident that w-e must be careful in drawing con- 

 clusions from experiments made under these conditions. 

 That wooden bed, however carefully made, would not be 

 I if the shape that nature would have given it, and any 

 deviation from nature's shape would cause unnatural 

 currents. It, however, does seem probable that something 

 of the nature of Prof. Thomson's diagonal under-tow will 

 exist even in river-shaped beds. 



The whole question of the flow of water in river beds is 

 extremely complicated. This is evidenced by the contrary 



NO. 1989, VOL. yy] 



results of the observations of your correspondents. But 

 little consideration is sufficient to show that this must be 

 so ; the variables are so many. We have, for instance, 

 variations in the curvature of the bend, in the velocity 

 of the water, and in the formation of the bed of the river, 

 which we must remember is dug out and shaped by flood 

 water for flood water, and is but little altered as the river 

 falls in volume. Take, for instance, the case supposed to 

 be represented by Prof. Thomson's model. Here, with a 

 certain curvature and a certain velocity of flow, we can 

 easily imagine the formation of the diagonal under-tow. 

 But if we were to increase the velocity of the flow this 

 cross under-current would decrease and ultimately cease, 

 and when a certain relation of velocity to curvature was 

 arrived at we would get the conditions referred to in 

 Mr. R. D. Oldham's letter in Nature of November 21, 

 where he says : — " Sand and even pebbles may be thrown 

 up to the surface of the water near the outer bank of the 

 stream, and where the waters have overflowed the banks 

 pebbles may be found lying on the dry ground after the 

 flood has passed away." 



In most of the rivers I know which flow in gravel 

 beds, where they are constantly cutting away their banks, 

 the main flow is more sinuous at low level than when in 

 flood. At low level the main stream runs into the pools 

 at the bends on the deep or concave bank, and as the 

 deep sides of the successive pools are on opposite banks 

 of the river, the stream has to cross its bed between the 

 successive pools. While in flood the swiftest flowing part, 

 on the surface at least, is near midstream, but the forma- 

 tion of the bed at the place and the flow above and below 

 may alter this in some cases. After the flood has fallen, 

 the river bed it has shaped has an infinite variety of 

 forms at different places, and the flow of the water at 

 any part must be studied with reference to that particular 

 part, and to the part above which has determined the cross- 

 section and velocity of the water coming to it, and also to 

 the formation of the part below which determines its 

 escape. 



There is one very common type of flow which frequently 

 presents itself in varying forms in rivers which alter con- 

 siderably in volume from time to time. After the flood 

 has fallen the river becomes, so to speak, divided into 

 streams and pools. Over the shallows the water runs 

 rapidly, while in the pools it moves slowly and somewhat 

 irregularly. The streams coming into the pools flow next 

 the concave banks, and come into the pools with some 

 velocity, which is soon lost in the slower movements in 

 the pools. The streams thus lose their kinetic energy, 

 which is converted into potential energy, raising the level 

 of the water at the place where the stream loses its 

 velocity. From this part of the pool, in addition to the 

 stream flowing down the pool, a reverse current is 

 started which flow-s back on the inner side of the pool, 

 flowing to the upper end of it, where it curves round and 

 flows downwards alongside the main stream. Part of the 

 back current is no doubt due to the inflowing main current 

 causing an induced current, but it seems to be mainly due 

 to the loss of kinetic energy of the stream, causing a rise 

 of the level of the water where its velocity is destroyed. 



.As to the cutting and wearing away of the banks of 

 rivers, that is mainly the result of eddies formed by the 

 flowing water meeting with obstructions, such as stones, 

 tree roots and stems, inequalities in the banks, &c., or 

 even by water impinging on water. One of the deepest 

 pools in a river I observed was entirely cJug out of its 

 gravel bed by eddies produced by the main river meeting 

 a large tributary stream at right angles and mingling 

 their w-aters in turbulent eddies ; and it seems probable 

 that the excavation of the deep pools generally found at 

 the foot of waterfalls have been greatly aided by the eddies 

 formed by the falling water meeting the quieter water of 

 the pools. 



The common practice in this country of protecting the 

 banks of rivers by means of little piers or " tooks " to 

 throw the water off them, and into the middle of the bed 

 of the stream, generally results in failure, because the piers 

 cause eddies, and deep pools endangering the banks are 

 frequently dug out bv these eddies ; and while these piers 

 tend to throw the water to the other side of the channel, 

 yet the sloping bed throws it back and causes it to strike 



