Jan. 28, 1 886] 



NA TUKE 



295 



valves, which prevent the air passini^ inwards. There are always 

 one or more sides on which the wind does not blow, allowing 

 the foul air free egress from within out. Some of the school 

 buildings where this system has been introduced are having as 

 much air passed through them as will refill the rooms every 10 

 or 15 minutes. 



This system, as explained, can be seen in operation at the 

 chemical laboratory of the Dundee University College, the 

 Harris Academy, Dundee, and at the Dundee High Schools, 

 the directors of which are introducing the system into another 

 large new school for girls, which is to be opened in a few 



months. WlLLL^M CUiNNINGHAM 



Dundee, January 12 



A Family of Rare Java Snakes 



At the Zoological Gardens, on Saturday, the gih inst., a 

 rather rare "Green -Tree Snake" {Dyyiophis prasi?m), from 

 Java, produced eight snakelings under circumstances which tend 

 to confirm recent observations regarding the uncertain period of 

 gestatii/n in snakes, otherwise the voluntary retention or deposi- 

 tion of their eggs or even their young. The mother was brought 

 to the Reptihum five months ago (August 15), and allowing two 

 months for her transportation from Java, it must be at least 

 seven months since she was captured and separated from her 

 mate. The normal period of gestation in a snake of this size 

 may be about three months, but incubation, which begins at once, 

 would in all snakes seem to depend a good deal on temperature 

 and on other propitious circumstances ; nor can it be positively 

 asserted that such or such a species is invariably oviparous or vivi- 

 parous, as in several instances the same snake has been known 

 to be both — i.e. under certain conditions an oviparous snake has 

 become viviparous. In sunny weather a high temperature is 

 obtained in the cages where this snake is ; and it is probable 

 that the late cold season may have materially affected this 

 Dryiophis. It is probable that, lacking the dense foliage of her 



! native forests, together with these adverse conditions of her 

 small glass dwelling, she retained her progeny until the latest 

 moment. 



The snakelings average 20 inches in length. The mother is 



I over 5 feet, and like all the family of whip-snakes is exceedingly 

 slender, with the long tail t.apering to a cord-like fineness. She 



I is of a bright emerald green, while the little ones are of a dull 

 ashy hue, with tongues of the same colour ; the mother's tongue 

 is pinkish. The parent has fed well on small lizards during her 

 captivity, but it is to be feared that the little family will fare 

 badly, as at the present lime suitable food is difficult to procure. 

 They were at once removed into another cage, or their mother 

 might have reduced their numbers at dinner-time. They soon 

 found their way to the water-pan and drank freely, and began to 

 cast their skins at an early day. 



Catherine C. Hopley 

 15, Queen's Crescent, Haverstock Hill, N.W. 



Vibration of Telegraph-"Wires 

 I NOTICED to-day a curious vibration of telegraph-wires near 

 here, and perhaps some reader of Nature may be able to 

 explain it. Each wire was vibrating rapidly, but instead of the 

 nodes being only at each post, there were several in each 

 span (of about 88 yards). The number of nodes varied in 

 «aeh span ; I counted seven in one, nor did the wires vibrate 

 together as a rule. In soir.e spans four out of five wires were 

 vibrating, and in others only one. The total amplitude of 

 vibration did got exceed I J inches, I should think. I noticed 

 this peculiar action in some five or six contiguous spans only. 

 There was a very hard frost at the time, and the wires were 

 coated with snow which had fallen some thirty-six hours 

 previously. There was no wind, and the sun was just breaking 

 through a fog. The wire was galvanised iron. No. 8 B.W.G. 

 E. DE M. Malan 

 Howden, East Yorkshire, Januaiy 19 



HEREDITARY STATURE^ 

 T T will perhaps be recollected that, at the meeting last 

 ■'• autumn of the British Association in Aberdeen, I 

 chose for my Presidential Address to the Anthropological 



s to the Anthropological 



Section a portion of the wide subject of " Hereditary 

 Stature." My inquiries were at that time advanced only 

 to a certain stage, but they have since been completed up 

 to a well-defined resting-place, and it is to their principal 

 net results that I shall ask your attention to-night. 



I am, happily, released from airy necessity of fatiguing 

 you with details, or of imposing on myself the almost 

 impossible task of explaining a great deal of technical 

 work in popular language, because all these details have 

 just been laid before the Royal Society, and will in due 

 course appear in their Proceedings. They deal with ideas 

 that are perfectly simple in themselves, but many of which 

 are new and most are unfamiliar, and therefore difficult to 

 apprehend at once. My work also required to be tested 

 and cross-tested by mathematical processes of a very tech- 

 nical kind, dependent in part on new problems, for the 

 solution of which I have been greatly indebted to the 

 friendly aid of Mr. J. D. Hamilton Dickson, Fellow and 

 Tutor of St. Peter's College, Cambridge. I shall there- 

 fore quite disembarrass myself on the present occasion 

 from the sense of any necessity of going far into explana- 

 tions, referring those who wish thoroughly to understand 

 the grounds upon which my results are based, to the 

 forthcoming memoir in the Proceediiios of the Royal 

 Society, and to that amplified and illustrated extract from 

 my Address at .Aberdeen, accompanied by tabular data, 

 which appeared among the " Miscellanea "of the Journal 

 of this Institute last November. 



The main problem I had in view was to solve the fol- 

 lowing question. Given a group of men, all of the same 

 stature, whatever that stature may be, — it is required to 

 be able to predict two facts regarding their brothers, 

 their sons, their nephews, and their grandchildren, re- 

 spectively, namely, _/frj/, what will be their average height ; 

 secondly, what will be the percentage of those kinsmen 

 whose statures wall range between any two heights we 

 may please to specify ; — as between 6 feet and 6 feet 

 1 inch, 6 feet I inch and 6 feet 2 inches, &c. ? 



The same problem admits of another rendering, be- 

 cause whatever is statistically certain in a large number is 

 the most probable occurrence in a small one, so we may 

 phrase it thus : Given a man of known stature, and ig- 

 noring every other fact, what will be the most probable 

 average height of his brothers, sons, nephews, grand- 

 children, &c., respectively, and what proportion of them 

 will most probably range between any two heights we 

 may please to specify ? 



I have solved this problem with completeness in a 

 practical sense. No doubt my formula; admit of exten- 

 sion to include influences of a minor kind, which I am 

 content to disregard, and that more exact and copious 

 observations may slightly correct the values of the con- 

 stants I use ; but I believe that for the general purposes 

 of understanding the nearness of kinship in stature that 

 subsists betw-een relations in different degrees, the 

 problem is solved. 



It is needless to say that I look upon this inquiry into 

 stature as a representative one. The peculiarities of 

 stature are that the paternal and maternal contributions 

 blend freely, and that selection, whether under the aspect 

 of marriage selection or of the survival of the fittest, 

 takes little account of it. My results are presumably 

 true, with a few further reservations, of all qualities or 

 faculties that possess these characteristics. 



Average Statures. — The solution of the problem as 

 regards the average height of the kinsmen proves to be 

 almost absurdly simple, and not only so, but it is ex- 

 plained most easily by a working model that altogether 

 supersedes the trouble of calculation. I exhibit one of 

 these : it is a large card ruled with horizontal lines i inch 

 apart, and numbered consecutively in feet and inches, 

 the value of 5 feet S inches lying about half way up. A 

 pin-hole is bored near the left-hand margin at a height 

 corresponding to 5 feet Z\ inches. A thread secured at 



