March i, 1877] 



NATURE 



375 



regions into the higher latitudes, and the polar air brings the 

 less rotatory velocity of the polar regions into the lower lati- 

 tudes. The latter constitute the trade-winds, which move more 

 slowly than the earth's rotation, and consequently appear as an 

 atmospheric current fnm the east ; the lormer con>titute the 

 "counter-irades," which move more rapidly than the earth's 



tation, and appear as an atmospheric current from the west. 



The centrifugal torce of the "countertrades," as i hey circle 

 roui:d the poles, is the cause of the polar depression of the 

 barometer. 



The law of reaction makes it impossible for the earth's rota- 

 tion to be either accelerated or retarded by the winds, and 

 consequently the entire "torsional force" exerted by the winds 

 on the earth must, at any given time, be equal in the easterly 

 and westerly directions. 



I have now described in outline what theory shows that the 

 circulation of the atmosphere would be in the absence of watery 

 vapour and in the presence of the sun's heat and the earth's 

 rotation ; and observation shows that such is the actual circula- 

 tion on the laige scale, and not taking account of local dis- 

 turbances. Joseph John Murphy 



Old Forge, Dunmurry, Co. Antrim, February 23 



Halo round Shadow 



It is not uncommon for an observer, when looking at his own 

 shaJow on rough ground or turbid water, to see its head sur- 

 rounded by a halo, of which the brightest part is in contact with 

 the shadow. 



This phenomenon has often elicited notice, but as far as I am 

 aware has not before now been explained, nor do those who 

 have mentioned it seem to have observed that its appearance 

 depended on the nature of the surface receiving the shadow. 



The conditions necessary for the production of these halos 

 are — 



1. That the screen, as whatever the shadow is cast on may be 

 called, should not be a continuous surface, but a number of small 

 surfaces with intervals between them, each of Ihese small sur- 

 faces of course casting its own shadow on whatever happens to 

 be behind it. 



2. That the shadow should be at a considerable distance from 

 the observer. 



3. That the light should not fall very obliquely on the screen. 

 The first of these conditions only is essential, but the fulfil- 

 ment of the last two makes the phenomenon more marked. 



Rough grass forms a good screen, especially if, as in the 

 diagram, conditions 2 and 3 are fulfilled by the shadow being 



cast on one side of a valley, while the observer is standing on 

 the other. 



In the case of the shadow on turbid water, it must be remem- 

 bered that it is not the surface of the water which forms the 

 screen, but the particles suspended in it. 



The general explanation of these halos is this — 



From the observer's point of view the screen in the immediate 

 neighbourhood of the shadow of the head is seen in nearly the same 

 direction as it would be from the source of light. In this direc- 

 tion, therefore, each of the small surfaces of which the screen is 

 made up will hide its own shadow, but this will be true of no 

 other direction ; and the effect on the whole will be that the 

 screen will appear brighter close to the shadow of the observer's 

 head than elsewhere. 



To examine this rather more in detail, let M ao be a section 



of the ground passing through the observer at M and his shadow 

 at o. Let 



o'p = r o'mp = < 

 o' P M = a nyht angle. * 



Let w and v/ be the projections on o' p of the average breadth 

 of the sections of the small kurlaces made by the pi ne M A<?, 

 and the average distance between them respectively, and let h 

 be the average distance of each of the small surfaces from its 

 own shadow. 



Then the amount of light received from any space rdQ 

 (w -1- 7c/) may, cateris paribus, be taken without any great error 

 as a measure of the brightness of the one whose mean radius is 

 r, and whose breadth is w + v/ (e/0 being a small rotation of r 

 about o' m), and this will be proportional to w + w' — /i sin. /, 

 The decrease in brightness is proportional to A and sin. i, and 

 will reach a maximum when k sin. i = w, if w < id , or = «/ if 



Outside the circle defined by this value of i the brightness 

 will be sensibly constant, because the quantities of which w, a/ 

 and h are the average values have all manner of actual values, 

 even in a very small space. 



These expressions are only approximate, but they serve, as 

 well as the longer exact formulae, to show the general laws of 

 the phenomenon, Arnulph Malloch 



Meteor 



This evening, at close upon twenty minutes past six, as I was 

 walking in my garden towards the almost full moon (which was 

 very brij^ht), I observed a brilliant meteor pass from right to left 

 over, and very near, the moon's disc. It was visible for a distance 

 of about twice her diameter. From the amount of daylight, and 

 the extreme brightness of the moon, I judge this meteor to be 

 worth recording. C, M. Ingleby 



Valentines, Ilford, February 26 



Tape-worm of Rabbits 



So far as I am aware the only evidence in favour of the view 

 that Boihriocephali present no hydatid stage is that which has 

 been furnished by the researches of Knoch. To me it has always 

 seemed that this evidence is insufficient fully to overcome the 

 analogical probability that tape-worms of this genus resemble 

 tape-worms of other genera in passing through a hydatid stage — 

 and this notwithstanding the occurrence of a ciliated embryo. 

 However, in my previous letter I ought no doubt to have alluded 

 to the researches of Knoch, and should certainly have done so 

 bad my object in writing been other than it was, i.e., merely to 

 ascertain whether anyone had as yet taken the trouble to trace 

 the life-history of the rabbit's tape-worm. 



February 20 George J. Romanes 



M- 



A PROBLEM IN THE NATURAL HISTORY 

 OF THE SALMON. 



R. FRANK BUCKLAND, in giving evidence before 

 the Parliamentary Committee, which during last 

 session of Parliament inquired into the condition of our 

 oyster fisheries, stated that "a salmon (J Salmo salar) 

 does not breed every year, but every three years ! " On 

 being asked by a member of the Committee if he had any 

 proof of his averment, Mr. Buckland stated that, " he 

 had a great idea of it," but was deficient in proof. Before 

 examining this alleged fact in the life of the salmon, ad- 

 vanced by Mr. Buckland, it is proper that we should state 

 briefly what induced him to make known his idea. 



While illustrating the theory of oyster spatting, and 

 telling the Commissioners that all the individual oysters 

 on a scalp would not be found exuding their young at the 

 same time, however favourable for spatting the period 

 might be, Mr. Buckland also enunciated his opinion as to 

 the periods at Which salmon spawn. That gentleman 

 holds that only one of every six oysters on a scalp will be 

 found in a procreant state during the same season ; and, 

 by way of clenching his illustration, he said, "you never 

 get salmon always breeding the same year, they take time 

 to recover themselves, and so forth," This latter state- 



