436 



NA TURE 



[September io, 1896 



He now adds that he might as well " have used any of the 

 ■other cases collected by Mr. Darwin." It is not a very material 

 point, but I do not find thai Mr. Darwin makes any reference to 

 AVolls's theory in his discussion of correlation, nor do 1 see any in 

 the body of the si.\th edition of the " Origin of Species," though a 

 passage is quoted from Wells's paper at p. xi. of the " Historical 

 Sketch " which is prefixed to it. It had, however, independently 

 occurred to Mr. Darwin, and he discusses it in a somewhat 

 different connection in the " Descent of Man" (i. pp. 242-245). 

 He remarks: — "That the immunity of the negro is in any 

 degree correlated with the colour of his skin is a mere con- 

 jecture ; it may be correlated with some difference in his blood, 

 nervous system, or other tissues." And he concludes : — " I 

 endeavoured with but little success to ascertain how far it held 

 good." Elsewhere he gives cases to show that "differences in 

 colour are correlated with constitutional differences." But these, 

 though interesting, seem to me too obscure to found any definite 

 conclusion upon. And no attempt is made to show on what 

 material basis, subject to variation, the constitutional difference 

 depends. 



The correlation principle as originally defined dealt then with 

 obvious and measurable characters. It is extended by Prof. 

 Lankester's " suggestion " to what is obscure, maybe unknown, 

 and perhaps unknowable. In considering the probable utility 

 of any specific character we shall, if the extended principle be 

 accepted, be always open to the objection that we cannot show 

 that the character is not the outward and visible sign of some 

 unobservable internal peculiarity. But that is a position which 

 I do not think we are bound to accept till something more than 

 a hypothetical case has been established. 



To sum up : Mr. Darwin based the correlation principle on 

 what is concrete and tangible ; Prof Lankester extends it to 

 what is intangible and hypothetical. It is not a question of what 

 is " apostolic and orthodox," but of what is susceptible of reason- 

 able proof. 



As I do not propose to continue this discussion any further, I 

 ■will take the opportunity of saying that I think it is a matter for 

 regret that, as Prof. Lankester was present at the meeting of the 

 Koyal Society when Prof. Weldon's paper was read, he did not 

 deliver himself on that occasion of his somewhat belated criticism. 

 Prof. Weldon's work is of extraordinary interest, and one cannot 

 but admire the self-sacrifice with which such laborious investiga- 

 tions have been prosecuted. If they want a defence, I think the 

 following passage from the "Origin of Species" supplies it. 



" It may metaphorically be said that natural selection is daily 

 and hourly scrutinising, throughout the world, the slightest 

 variations ; rejecting those that are bad, preserving and adding 

 lip all that are good ; silently and insensilily working, whenever 

 and wherever opportunity offers, at the improvement of each 

 organic being in relation to its organic and inorganic conditions 

 of life. We see nothing of these slow changes in progress, until 

 the hand of time has marked the lapse of ages, and then so im- 

 perfect is our view into long-past geological ages, that we see 

 only that the forms of life are now different from what they 

 formerly were." (Sixth edition, pp. 65-66.) 



I do not myself see how the slow and ordinarily imperceptible, 

 but inevitable action of natural selection can be demonstrated 

 except by the statistical method. But, firmly as I believe in the 

 inevitableness of that action, I confess that the results attained 

 by Prof. Weldon surpassed my expectations. I am unable to 

 agree with Prof. Lankester, that the investigation does not 

 satisfy the canons of scientific inquiry. The hypothesis on which 

 it appears to me to be based is, that the mean configuration of any 

 organism at any moment is an optimum. In order to test that 

 liy the statistical method, the choice of measurements is a mere 

 matter of convenience. W. T. Thisei.ton-Dyer. 



Kew, August 29. 



Thermometer Readings during the Eclipse. 

 I STARTED on July 30 in the King Harold, and arrived at 

 Vadso on August 6. On board this vessel, ainongst others, were 

 Prof. Rambaut and Dr. Hugh R. Mill, of the Geographical 

 Society, who I see has sent a note which a]5pears in Nature of 

 August 27, as to some observations of temperature he took 

 during the eclipse. I was constantly with Prof. Rambaut on the 

 island at Vadso, and he particularly requested me to observe 

 the temperatures of sun, and shade thermometers during the 

 eclipse at the position he had taken for his observations, which 

 were specially directed to the degree of polarisation of different 

 parts of the corona. I enclose a diagram of my observations, 



NO. 1402, VOL. 54] 



which Prof. Rambaut has suggested I should send to Nature, 

 should you think they are worth recording. The fall of the sun 

 thermometer (which unfortunately was fully shaded by cloud) 

 was, from 4h. lom. to just after totality. 2", and its recovery 



clipse of the 



i-i shade thermometers during the' 

 .August g, taken at Vadso. 



from that point to 5h. 56m., last contact, was 3''6. The shade 

 thermometer showed greater variations, viz. a fall of 3°'35, and 

 subsequent rise at 5h. 50m. of 5°'6. 



H. WOI.I.ASTON Bl.AKE. 

 S Devonshire Place, W., September 3. 



Sailing Flight. 



Mr. Peai. (Nature, vol. liv. p. 317) having again brought 

 up this matter for discussion in the columns of N.vruRE, I 

 would like to make a new suggestion concerning it, which I 

 have long had on my mind. It will be remembered that Lord 

 Rayleigh (Nature, vol. xxvii. p. 534) assumed an increase of 

 wind-velocity with altitude to explain the facts of circular 

 soaring, and that quite recently Langley (Anier. Journ. Set'., 

 vol. xlvii. p. 41) has tried to explain the same phenomenon by 

 the assumption — supported in his case by direct observation — 

 that the velocity and direction of the wind is subject to great 

 and rapid changes. Concerning this latter statement, I must 

 say that although in a thunderstorm great irregularities can be 

 observed in the upper air-currents, the shape and relative con- 

 stancy of small clouds in fine weather seem to show that under 

 ordinary conditions the upper air-currents are much steadier 

 than Langley assumes, and that, therefore, soaring birds can by 

 no means always depend on the presence of wind-irregularities 

 sufficiently great to sustain them. Although no doubt wind- 

 velocities generally increase with altitude, I do not believe 

 that such an increase will always be present, nor that it will, 

 when present, be usually sufficiently great to produce the 

 force necessary for raising a bird. We observe, however, that 

 birds do soar nearly always, perhaps even more frequently in 

 fine weather, when the currents are more steady, than in rough 

 weather, when they are more irregular. 



Under these circumstances it seems to me that neither Lord 

 Rayleigh's nor Langley's assumptions concerning the source 

 from whence these birds derive the power of overcoming gravity 

 can be correct. It seems to me, doubtless, that a steady hori- 

 zontal wind of equal velocity in different altitudes does enable 

 them to soar and to rise. It is remarkable that this soaring 

 without loss of elevation is always accompanied by circling. 

 Elevation is not known to occur without circling, as it might 

 if Langley's views were correct. Were the bird attached to 

 the earth by a string like a kite, it could be and, if the wing- 

 planes were placed in proper positions, would be sustained and 

 raised by a (lurely horizontal and steady wind. Now it seems 

 to me that the circling replaces the string. A circling top retains 

 its position on account of the force in its rapidly circling parts. 

 Could not the soaring bird produce — through circling — a similar 

 stability which, acting like a kite-string, would enable it to 

 oppose itself to the wind, and thus convert the horizontal 

 wind-force partly into a vertical, lifting force ? Mr. Peal, in 

 his last letter (/.I.) very correctly remarks that the connecting- 



