October 20, 1898] 



NA TURE 



593 



Now from this process of segregation it must have happened 

 that when "dextro- and l?evo-protein were simultaneously 

 formed" the two kinds of molecules, differently related to 

 environing actions (say ethereal undulations alike in nature and 

 direction), separated themselves into groups of their respective 

 kinds. It is true that in virtue of the small differences between 

 the two classes of molecules, the minute differential actions of 

 forces upon them might be long in producing their effects ; and, 

 further, that the segregation might be impeded by restraining 

 forces. But when we remember that segregations take place in 

 long periods of time even where the restraining forces are very 

 great, as instance the formation of hematite nodules and flints in 

 chalk-formations or of siliceous concretions in limestone, the 

 implication is that the segregation would slowly, if not quickly, 

 take place. And then the molecules of either group would 

 exhibit just that optical activity which Prof. Japp, following 

 Pasteur, alleges can result only from molecules formed by vital 

 action. 



I do not draw attention to this truth for the purpose of show- 

 ing the adequacy of the physico-chemical interpretation of life, 

 but for the purpose of showing the inadequacy of Prof. Japp's 

 argument against it. My own belief is that neither interpret- 

 ation is adequate. A recently-issued revised and enlarged edition 

 xaf the first volume of the "Principles of Biology" contains a 

 chapter on " The Dynamical Element in Life," in which I have 

 contended that the theory of a vital principle fails and that 

 the physico-chemical theory also fails : the corollary being that 

 in its ultimate nature Life is incomprehensible. 



Brighton, October 12. Herbert Spencer. 



Organic Variations and their Interpretation. 



I SHOULD like, if you will kindly afford me a little space, to 

 offer a few remarks on Prof. Weldon's presidential address to 

 Section D of the British Association. 



The first part of that address deals with the question whether 

 individual variations are fortuitous, i.e. occur by chance. It 

 contains a ver>- able and lucid exposition of the fact that the 

 distribution of individual variations is of a similar kind, and is 

 open to the same mathematical treatment, as events which 

 happen by chance. I do not think that any one has denied 

 this. It does not admit of dispute. But it is no answer 

 whatever to the reasoning of those who oppose the theory 

 of Natural Selection. The question is whether a given 

 modification, the degrees of which are distributed among 

 individuals according to what may be called the law of chance, 

 originated accidentally, or as the result of a definite ascertainable 

 cause. To give an illustration. If I plant a hundred 

 or a thousand sunflower seeds in good soil in a market 

 garden, at about equal distances from one another, I get 

 a number of sunflower plants which will not all be of the 

 same size. If I measure their heights, or take their 

 weights, I shall find that these magnitudes are so distri- 

 buted as to form one of Prof. Weldon's curves. If I take 

 another hundred or thousand seeds from the same sack, and 

 plant them in flower-pots, each 6 inches in diameter and of 

 exactly the same capacity, placing the flower-pots in the same 

 garden, I shall get a number of sunflower plants whose heights 

 or weights will form a curve of the same kind. But the mean 

 height or weight of the second lot of plants will be very much 

 less than that of the first lot. This I know to be true because 

 I have tried it. The distribution of the magnitudes has nothing 

 whatever to do with the cause of the difference in the two cases. 

 That cause is limited nourishment in the second case. Similarly 

 in the progressive modification of animals and plants under 

 natural conditions, the distribution among individuals of the 

 degrees of a character has nothing whatever to do with the 

 question of the cause of the character. When selection takes 

 place, by breeding from the larger or the smaller variations, the 

 mean of a character may be raised or lowered, but the question 

 is whether this can be done without regard to conditions of life 

 or not. In numbers of cases there is reason to believe that it 

 cannot. And there is reason to believe that in numbers of 

 cases the mean of a character can be raised or lowered by the 

 application of definite conditions without any selective breeding 

 at all. I will not attempt to prove this here ; all I wish to 



reason that the word conservation is doubly inappropriate. Conservation 

 connotes a conserver and an act of conserving — conceptions utterly at 

 variance with the doctrine asserted ; and it also implies that in the .absence 

 of a conserver and an act of conserving, the energy would disappear, which 

 i> also a conception utterly at variance with the doctrine asserted. 



NO. 15 I 2, VOL. 58] 



point out is, that Prof. Weldon's argument does not touch the 

 question. 



Still more serious objections must be made to Prof. Weldon's 

 evidence concerning the actual operation of selection with 

 regard to .the frontal breadth of Carcinus tnanas. I do not 

 dispute his measurements, but his interpretation of them, 

 which seems to me obviously and demonstrably unsound. He 

 finds that the mean frontal breadth of the crabs at Plymouth 

 was less in 1895 than in 1893, and less in 1898 than in 1895. ^ 

 have always held that he courted failure by taking for investiga- 

 tion a character which is known to be undergoing progressive 

 change in the individual during growth. We know that change 

 in the proportions of a crab occur only at the ecdysis. It is, I 

 think, certain that the number of ecdyses depend on age, not on 

 size. Prof. Weldon himself remarks that the estimate of age by 

 size is a dangerous proceeding. Yet for individual variations he 

 compares crabs of the same size, not of the same age. Now the 

 results he finds with regard to the diminution in frontal breadth 

 in terms of total length would be exactly the same, if the growth 

 of the crabs had been less in 1895 and 1898 than in 1893 ; in 

 other svords, if the crabs of the same size had been in these 

 years, on the average older, had on the average passed through 

 more moults ; for the older crab has a relatively smaller frontal 

 breadth. Now have I any reason for supposing that the crabs 

 grew more rapidly in 1893, and do I suppose that the increased 

 muddiness of the water in Plymouth Sound caused a diminution 

 in the rate of growth ? I do not suppose that mud had anything 

 to do with it, but I have good reason for holding that 

 crabs, like oysters, grow faster and larger when the water 

 is warmer. Here is what Mr. Garstang wrote in 1894 

 concerning the summer of 1893 = " Under the influence of 

 the great heat the temperature of the Channel waters rose con- 

 tinuously, until in August it attained a point unprecedented for 

 a quarter of a century ; and it was of the highest interest to 

 observe the effect of this high temperature, and of the prolonged 

 calmness of the sea, upon the floating population of the neigh- 

 bouring portion of the Channel. Numbers of semi-oceanic 

 forms which rarely reach our shores arrived in remarkable pro- 

 fusion. In June the tow-nets were crowded with Salps, while 

 towards the latter end of August they were almost choked by 

 masses of living Radiolaria." The beginning of the year 1895, 

 on the other hand, was exceedingly cold, and in the summer the 

 water temperature was less than in 1893. I' i^ not certain, as 

 so few crabs were measured in 1898, whether their mean frontal 

 breadth was really less than in 1895. But it is a fact that although 

 last winter was unusually mild, the water temperature off Looe 

 in May and June was lower than in the same months in 1895. 



It is remarkable that Prof. Weldon found the change in 

 female crabs was less than in males, and it is difficult to under- 

 stand why the sexes should be affected in different degrees by 

 an increase in the muddiness of the water. On the other hand, 

 as the males in crabs generally are larger than the females, the 

 former would necessarily be more afifected in their growth by 

 temperature. 



Next as to the experiment which is adduced to show that the 

 increase of sediment is the cause of the selective destruction of 

 the crabs with greater frontal breadth. The survivors of crabs 

 placed in water with china clay had narrower foreheads. But 

 this merely means that they were on the average older, and the 

 younger specimens were killed first, which is what might be 

 expected. Prof. Weldon believes that the cause of death was 

 the entrance of the sediment into the gill chamber, but it 

 appears that the dead crabs had been in the muddy water, while 

 the living were killed after removal. There is no evidence that 

 the clay entered before death, and any dead crab which had 

 been some time in muddy water constantly stirred would 

 probably have mud. on its gills. 



Lastly there is the experiment of keeping crabs in bottles for 

 a period including a single moult. At first the crabs with 

 broader foreheads died, and in this case the death is attributed 

 to the putridity of the water. In this case there was no 

 sediment, and putrefaction in the water has the same effect as 

 sediment, a fact perfectly in agreement with the view that under 

 unfavourable conditions the younger die first, but inconsistent 

 with the view that death is due to the greater filtering power of 

 the branchial apparatus in the narrower-fronted crabs. The 

 mean frontal breadth having been decreased by selective deaths 

 before the moult, was found, after the moult, to be greater than 

 that of crabs from the sea of the same size. This again is easily 

 explained by increased growth. The crabs in the bottles were 



