350 



NATURE 



[January 3, 19 18 



which is given a clear and simple account of the uses 

 of food, together with suggestions for the making up 

 of dietaries, having regard to both the economic and 

 patriotic aspects of food problems in war-time. The 

 pamphlet is published by the Department of Agricul- 

 ture and Technical Instruction for Ireland, and can 

 be had free on application. 



A SHORT paper by Mr. E. S. Goodrich in the 

 Quarterly Journal of Microscopical Science {vol. Ixii., 

 part 4) throws considerable light on the interesting 

 problem of the homologies of the coelomic spaces in 

 various groups of the animal kingdom. The "proboscis 

 pores," which lead from the proboscis coelom to the 

 exterior in Balanoglossus, are believed to be 

 represented in Echinoderms by the water pore, 

 and in Amphioxus by the opening of Hat- 

 schek's pit, as originally suggested by Bateson. 

 But the anterior coelomic sacs of Amphioxus, 

 one of which becomes metamorphosed into Hatschek's 

 pit, are homologous with the premandibular somites 

 of Craniates. The cavities of these somites may de- 

 velop a tubular connection with the hypophysis, which 

 is to be identified with the proboscis pore of lower 

 forms, while the hypophysis itself is the homologue of 

 the ciliated "wheel organ" in the buccal cavity of 

 Amphioxus. These views are supported by an interest- 

 ing reconstruction of part of the head of an embryo 

 torpedo. 



In the Transactions of the Royal Society of South 

 Africa (vol. vi., part i, 1917) Mr. P. A. Wagner pub- 

 lishes an exhaustive monograph on the national game 

 of skill of Africa. The game, in one form or the 

 other, is played in rows of holes scooped out of the 

 ground, or on wooden, stone, or even ivory boards. 

 As a matter of fact, it is not* confined to Africa, being 

 played in Syria, Arabia, Bombay, Ceylon, the Malay 

 Peninsula, and along the entire southern coast of Asia as 

 far as the Philippine Islands. It is essentially a war 

 game, two players, or sides, directing a contest between 

 armies of equal strength, the object being the capture 

 or "killing" of "men," who are represented by small 

 stones, seeds, shells, or fragments of dried cow-dung. 

 It is often played for a stake, but it is certainly not a 

 gambling game, as some writers have maintained. It 

 is also incorrect to say that it is very intricate, though 

 it does require a certain facility in ready reckoning. 

 It is of considerable antiquity, being known to the 

 Arabs of the Middle Ages, and stone boards and frag- 

 ments of others have been found in the neighbourhood 

 of ancient ruins in Rhodesia. The problem of the 

 methods by which this game appears in such a wide 

 area is interesting, but is not dealt with by Mr. 

 Wagner. 



In a study of the natural regeneration of the Douglas 

 fir and other conifers in the Pacific coast forests of the 

 United States, published in the Journal of Agricultural 

 Research, vol. xi., pp. 1-26 (October, 1917), J. V. Hof- 

 mann shows that when a large area is either burnt or 

 cut away, the complete restocking which usually takes 

 place does not result from the seeds that are scattered 

 by surviving trees on the area or in its vicinity. The 

 distance from the parent tree to which seed is carried 

 by the wind is very small, 150 to 300 ft. Consequently, 

 if only wind-dispersed seed germinated, the regenera- 

 tion of a large area would not be completed until after 

 the growth of several generations of trees. The repro- 

 duction is never a gradual creeping out from surround- 

 ing bodies of green trees, but is a sudden taking posses- 

 sion of the whole area bv a dense growth of seedlings. 

 The regeneration is really effected by the seed which 

 is stored in the ground amidst the litter and humus, 

 NO. 2514, VOL. 100] 



which are not destroyed in the swift passage of the 

 ordinary forest fire. The litter is found on examina- 

 tion to contain a large number of germinable seed. 

 The ordinary form of succession is the replacement of 

 the forest almost immediately by the same species 

 as composed the original stand, and usually in the 

 same proportions. This paper is well illustrated with 

 diagrams and photographs. One plate is a view of the 

 reproduction on the Yacolt " Burn " of 1902 in the 

 Columbia National Forest. The extent devastated by 

 fire is 604,000 acres. No green trees are visible, yet 

 there are seedlings growing among the snags over the 

 whole area. 



Prof. P. L. Mercanton, in the Revue gdndrale des 

 Sciences for November 30 last, discusses the results of 

 the more recent observations on the advance and re- 

 treat of glaciers, especially Alpine. Systematic work 

 was begun by Prof. Forel thirty-seven years ago in the 

 Swiss Alps, and for at least thirty of them the move- 

 ments of the Rhone Glacier and the two at Grindelwald 

 have been carefully noted. Those on the northern side 

 of Mont Blanc have also been studied, and similar 

 work is now being carried on in other icefields. But 

 the main advances and retreats of those glaciers and a 

 few others in the Alps are known for fully three 

 centuries, and estimates of their periods have been 

 attempted. These do not correspond with Wolf's 

 eleven-year period, or with the thirty-five-year one of 

 Bruckner. Some causes affect their movements other 

 than the snowfall in the upper region — that of the n^v6 

 — ^^and the ablation due to temperature changes in the 

 lower; for of two adjacent glaciers, one may be ad- 

 vancing while another is retreating. Recent observa- 

 tions, as Prof. Mercanton points out, indicate that the 

 volume and the length of a glacier can to some extent 

 vary independently, or, in other words, that the ice 

 moves down a valley from the more expanded n^v^ 

 basin at its head, not with perfect uniformity, but with 

 local intermittence, so that a belt near the end may 

 be swelling up in a wide mound, and thus the actual 

 volume of ice be increasing, while the end itself is in 

 retreat. Evidently, as Prof. Mercanton observes, the 

 subject of glaciers and their history is not yet ex- 

 hausted. 



Prof. Filippo Eredia has recently published in the 

 Bollettino d'Informazione of the Italian Colonial Office 

 a useful note on the frequency of snow in Tripoli and 

 in Algeria. In the last-named country at sea-level 

 snow is rare, since only one fall in the whole year 

 may be expected. At a height of 600 metres six 'falls 

 per annum occur on the average, while at double this 

 elevation twenty-five falls are experienced. In Algeria 

 and Tunisia the most frequent and extensive snow- 

 falls occurred in the winter of 1890^1, while 1884, 

 1904-.:;, 1913, and 1915 were also characterised by 

 abundant snowfalls. Some interesting photographs 

 are given of snow scenes in Tripoli during the snow- 

 storms of February, 1913, and February, 1915. 



^ An interes;ing instance of the way in which the solu- 

 tion of a problem in one branch of science provides or 

 aids in the solution of a problem in an entirely different 

 branch was brought before the Institution of Mechan- 

 ical Engineers on December 14 by Messrs. Griffith and 

 Taylor in a communication entitled "The Use of Soap 

 Films in Solving Torsion Problems." The authors 

 show that the equations which determine the stress in 

 a rod of any section subjected to twist are identical 

 with those which determine the slope of a soap film 

 formed in a hole of the same shape as the section, in a 

 horizontal plate above which it protrudes owing to 

 a small excess of pressure on the under-side of the 



