30 



NATURE 



[May 14, 189: 



propagation of each with its own kind while preventing crossing ; 

 but numerical infertility of first crosses produces what I call 

 negative segregation, for, though it is unable to secure segregate 

 breeding, it lessens or obliterates the swamping effects of any 

 crossing that takes place, and is therefore of great importance in 

 the preservation of distinct varieties and species when the posi- 

 tive forms of segregation only partially prevent crossing. The 

 four forms of hybrid inferiority mentioned above are also forms 

 of negative segregation, and, though of the highest importance 

 when co-operating with prepotential segregation or any other prin- 

 ciple that partially prevents cross unions, are, it seems to me, 

 incapable of preserving distinct varieties or species, when un- 

 assisted by any degree of positive segregation. 



We are now prepared to see how the different meanings of in- 

 fertility have occasioned more or less misunderstanding in the 

 discussion of physiological selection and its effects. With Dr. 

 Romanes, the seven forms of segregation above-mentioned are all 

 forms of infertility between races, and therefore are all causes of 

 physiological selection ; while in my nomenclature, all but the first 

 are included under impregnational segregation, and only the 

 second and third are considered forms of cross infertility (or, in 

 other words, of segregate fecundity). Using the term in this 

 restricted sense, I have elsewhere maintained that it is very im- 

 probable that cross infertility is, in any case, the only isolative 

 principle securing the continuance of distinct varieties and 

 species indiscriminately commingled on the same area, even 

 when the elements are freely distributed ; and as this statement 

 is liable to be taken as equally applicable to physiological selec- 

 tion, I wish to have it clearly understood that, in my usage, the 

 two terms are not equivalent, and what I have said of cross in- 

 fertility is not in the same sense true of physiological selection. 



In Dr. Wallace's criticism of physiological selection, he seems 

 to limit the meaning of infertility between races to numerical in- 

 fertility of first crosses, and then assumes that this is the only 

 incompatibility that is included under physiological selection. 

 This limitation,- if correct, would of course limit the effects that 

 could properly be attributed to this principle. 



Before closing I wish to raise the question whether a high 

 degree of selective numerical fertility between races is not 

 always associated with some degree of selective potential fer- 

 tility. Or, using infertility in the more restricted meaning given 

 in my nomenclature, is not a high degree of segregate fecundity 

 and cross infertility always associated with some degree of segre- 

 gate prepotence and cross impotence ? As we know that these 

 two forms of incompatibility are usually, if not always, associated 

 in the segregation of species, is it not probable that they are simi- 

 larly associated in the segregation of varieties? Again, as we know 

 that segregate prepotence, when associated with the free distrit)u- 

 tion of the fertilizing elements, will produce prepotential segrega- 

 tion, effectually preventing crossing, without impairing powers of 

 survival, and as there are many cases in which the continued 

 segregation of varieties occupying the same area is due entirely 

 to this principle, and many other cases in which it is due to 

 weakened forms of this principle associated with other forms 

 of incompatibility in the reproductive system, and still other 

 numerous cases in which partial isolation (pioduced by a slight 

 diversity of habits, or by the occupation of adjoining districts) 

 would be speedily broken down except for these physiological 

 incompatibilities, are we not fully warranted in the assertion 

 that physiological selection is an essential factor in the evolution 

 of many species ? 



The importance of this form of segregation having been 

 recognized, the question naturally arises as to what have been 

 the causes through which the incompatibility has ceased to be 

 sporadic, and has become racial. As Dr. Romanes has not 

 entered on the discussion of this point, I have given the more 

 attention to it. I think I have succeeded in showing : (i) that 

 any portion of a species subject to temporary isolation, through 

 occupying a new station or district, is more or less liable to 

 become incompatible with the rest of the species, owing to the 

 cessation of reflex selection, by which the mutual fertility and 

 other compatibilities of an inter- generating stock are kept in 

 force (see Nature, vol. xlii. pp. 28 and 369) ; (2) that partially 

 segregative endowments are, through the very laws of propaga- 

 tion, cumulative (see " Divergent Evolution," Linn. Soc. Journ. 

 — ZooL, vol. XX. pp. 246-260) ; (3) that all the transformations 

 that arise in forms thus segregated are inevitably divergent, and 

 not parallel (see " Intensive Segregation," Linn. Soc. Journ. — 

 Zool., vol. xxiii. pp. 312-322). John T. GULICK. 



26 Concession, Osaka, Japan. 



Propulsion of Silk by Spiders. 

 The author (" O, P. C") of the article on " Arachnida" in. 

 the "Encyclopedia Britannica," says:— "The emission of silk 

 matter appears to be a voluntary act on the part of the spider ;. 

 but it is a disputed question among arachnologists whether 

 spiders have the power forcibly to expel it, or whether it is 

 merely drawn from the spinnerets by some external force or 

 other. Mr. Blackwall, author of the ' History of Spiders in 

 Great Britain and Ireland,' is of the latter opinion. Mr. R. U. 

 Meade (Yorkshire) in Report of the Briti!^h Association, 1858, 

 thinks, that (from microscopic anatomical investigations which 

 he has himself made) there is good evidence of spiders having 

 the power to expel it ; for he finds a certain muscular arrange- 

 ment which would apparently suffice to give this power, and 

 observers have actually seen the lines propelled." 



Owing to the doubt herein expressed, may I ask your inser- 

 tion of a chance observation lately made by me upon a spider, 

 which has convinced me of the truth of the theory that spiders 

 do expel their lines at will, and this, too, as secondary to one 

 still remaining attached to the spinnerets ? 



She was hanging from the ceiling about 3 feet from a 

 mullioned window, against which I was able to observe her 

 movements most accurately. I was first led to observe her 

 closely, by finding myself attached to her within one minute of 

 my approach. On my breaking this line, she attempted to 

 regain the ceiling ; a breath of air from me stopped and brought 

 her down again, when I saw her draw her legs together, pull 

 her head up higher than the spinnerets of her abdomen by 

 means of her ceiling-line, md, following upon no visible effort oj 

 hers whatever, I was the next moment conscious of the presence 

 of another line stretching out from her spinners to a distance 

 short of 3 feet, and at an angle of about 75° with the first. This 

 line failing to find an attachment floated upwards and lay along- 

 side of the other, and the spider again made for the ceiling. 

 Nine times during the space of one hour, I got her to repeat 

 this attempt to make a horizontal connection. Between two of 

 the intervals of her attempts, I called in two naturalist friends 

 who both witnessed with me, and at the same instant of tune, 

 the sudden appearance of the new line. 



With each successive trial, I was able to substantiate and 

 improve my observation ; at first the appearance of the line 

 seemed instantaneous, as to its whole length ; next I was able 

 to detect its elongation of its-elf after about 2 feet of its length 

 was visible ; then I could see it leaving the spinnerets ; and 

 finally, during the last moment of its travel, I could perceive 

 very distinctly that it drew the spider slightly forward. 



From these premises I can but infer that the viscid matter 

 contained by the silk-glands, which, at the ordinary slow rate of 

 emission, turns to gossamer immediately upon its exposure to the 

 air, when expelled as now, violently, remains viscid sufficiently 

 long to reach a certain distance. 



These secondary threads, carried towards the ceiling by the 

 spider, were never brought down again when she fell to the 

 length of the main line, but were each time left, disconnected 

 from her, at the spot from whence she fell when I ble\y her. 

 Their loose end invariably floating upwards until alongside of 

 the spider's main line, was, I think, noteworthy. 



In conclusion, I would say that sight seemed to play no part 

 in her choice of a direction for the connecting line ; though I 

 was close to her all the time, and indeed the only object appar- 

 ently which was close enough, she only hit me the first time, 

 when perhaps she had heard my approach ; this may strengthen 

 the remarks made by Mr. C. V. Boys in your number for 

 November 13, 1890, where he says: ". . . . sight, as we 

 understand the term, in spite of their numerous eyes, seems to 

 be absent." S. J. 



St. Beuno's College, St, Asaph, N.W., April 27. 



The Crowing of the Jungle Cock. 



I THINK there can be little doubt that Mr. H. O. Forbes has 

 fallen into the same mistake as I had, in regard to Mr. Bart- 

 lett's statement that "none of the known wild species are ever 

 heard to utter the fine loud crow of our domestic cock." 



At first I look this to mean that the jungle cock did not crow 

 at all, and was collecting notes from sporting men here, to 

 supplement my own 26 years' experience, when yours of 

 February 5 arrived, and by it I see that Mr. Barllelt implies 

 that the crow is not so full, loud, and long, as that of our barn- 

 door cock. 



NO. I I 24, VOL. 



44] 



