Sept. 13, 1883] 



NATURE 



477 



the error will at sorne future time be found to fall within this 

 limit. The law of probabilities requires this, if the maximum 

 limit falls within 1". But who is prepared to select a particular 

 star and say tint the absolute position of this star in space can- 

 not be more than o'2" in error ? 



<'. At present an arbitrary hypothesis is necessary in the dis- 

 cussion of ihe problem. Airy assumed that the relative distances 

 of the stars are proportional to their magnitudes ; and he found 

 slightly different results according to different nudes of treat- 

 ment. Safford assumed that the distances are, at least approxi- 

 mately, in inverse proportion to the magnitude of the proper 

 motions. The general result of his investigations up to this 

 point is that there is some hope of using the solar motion as a 

 base to advance our knowledge of stellar distances. Later inves- 

 tigations have been made by De Ball, but the details have not 

 yet come to hand. It is understood, however, that his results 

 coincide in a general way with those previously obtained. 



It is clear from this brief review that we have here a field of 

 investigation worthy of the highest powers of the astronomer. 

 The first step has been taken in the survey of the heavens 

 carried on under the auspices of the Gesellsehaft. It remains 

 for the astronomers of the present generation to solve the diffi- 

 culties which now environ the problem, and prepare the way fur 

 a more perfect scheme of observation in the next century. 



INDIAN METEOROLOG \ ' ' 

 III. 

 THE next paper we shall notice is No. IX., by Fred. Cbam- 

 ■*■ bers, on " The Winds of Kurrachee." The station dealt with 

 is not only a representative one of the Arabian sea current, but 

 is remarkable for exhibiting the highest average monthly wind 

 velocity of any place in India. The observations used were 

 furnished by a Beckley's anemograph for 1873, 1874, and 

 1S75.- 



In discussing the annual variation, Mr. Chambers adopts a 

 plan which has been followed out with much success by his 

 brother in his great work on the meteorology of the Bombay 

 Presidency, viz. its separation into nortnal and abnormal north 

 and east components. 



It is thence found that the former are closely related to the 

 corresponding barometric variations, and represent that part of 

 the grand monsoon system which affects Kurrachee, while the 

 latter are found to be connected with a system of local convec- 

 tion currents, due to (relatively) local temperature variations. 

 These latter, though subordinate to the former in point of mag- 

 nitude, are still sufficiently large to mask the true nature of the 

 regular monsoon currents which obey the barometric law. This 

 is more especially the case in Bengal, where, as it appears both 

 from evidence furnished in this paper and elsew here, the activity 

 of the monsoon currents is far less than on the west coast of 

 India, while the absolute efficiency of the local variations is 

 about the same. 3 



Another important; result deduced, is that the causes which 

 produce the abnormal variations in the wind and pressure com- 

 ponents, are similar to those which produce the annual variations. 

 Thus, when the barometer rises abnormally a tenth of an inch, 

 it is accompanied by an abnormal wind of 4^4 miles per hour 

 from N. 55° E., while a similar rise in the barometer from 

 summer to winter gives rise to a wind of 47 miles per hour 

 from N. 57° E. 



This principle, which, though a priori probable, has not 

 hitherto been supported by direct evidence, is without doubt 

 destined to play an important part in the meteorology of the 

 future, and to form one of the few channels by which we may 

 hope to arrive at a correct knowledge of the effects of the sus- 

 pected intrinsic variation of solar radiation on terrestrial meteor- 

 ology. Thus Mr. Chambers says : "If the sun's heat is itself 

 subject to fluctuation*, either periodical or irregular, corres} 01. d- 

 ing meteorological effects similar to those which are produced by 

 the sun's change of position must result ; " and he adds : " The 

 relation at Kurrachee appears to be one of the kind that would 



1 Continued from p. 433. 



3 The small elevation of the anemograph (only iy6 feet above the ground) 

 is open to some objection, but this is a good deal compensated for by its 

 unusually free exposure. 



3 The resultant ranges of the wind variations obeying the barometric law 

 are as follows : — 



Kurrachee 26*6 



Bombay 20*5 



Calcutta 6'a 



be anticipated on the supposition of the sun's heat being variable, 

 and in itself affords a reason for suspecting, if it does not tend to 

 prove, such variability." 



In discus.-ing the diurnal variations, Mr. Chambers divides the 

 winds into two great cla-ses, convection or ordinary current-', in 

 which the air moves from relatively cool to relatively warm 

 regions, and anti-convection currents, or "winds of elastic expan- 

 sion " as Blanford calls them, which blow outwards from regions 

 of high temperature. Each of these classes is again divisible into 

 two sub classe=, (1) general and (2) local. 1 



If each of these systems is possible, as Mr. Chambers infer , 

 the resultant variation is evidently a very complex one, and the 

 main difficulty in discussing it, evidently consists in being able to 

 adequately separate each component in turn from the rest. Fi 

 this purpi se Mr. Chamber- employs Bessel's formula, and though 

 he admits that the components derived by this method, do not 

 necessarily represent physically distinct variation-, its use in this 

 case, as well as in others throughout this work, is attended with 

 such favourable results, as to constitute a plea in favour of its 

 more general adoption by English meteorologists. 



To follow all the details of the investigation would be beyond 

 our scope. It may therefore be briefly noted that the greater 

 part of the variation of the north component, is due to the alter 

 nate land and sea breeze (convection currents), while a portion 

 at any rate of the variation of the east component, is due to local 

 anti-convection currents which prevail only in the drier months. 

 Further, the direction of the local anti-convection currents varies 

 with the varying position of the centre of maximum temperature 

 range in the peninsula, w hile that of the coast convection currents 

 is nearly constant. 



By an ingenious plan for eliminat'ng the variations due to 

 coast convection currents, and by choosing the months so as to 

 reduce the local anti-convection currents to a minimum, the 

 existence is further proved of a system of general anti-convection- 

 currents, which, it may be remarked, were first 1 noticed by Mr. 

 Laughtmin 1S71, consisting of a double diurnal oscillation of 

 the east component, which in the case of Kurrachee reaches it- 

 maxima at 10 a.m. and 9 p.m. and its minima at 4 p.m. and 

 2 a.m. respectively. These general anti-convection currents have 

 been likewise proved by Mr. Chambers to exist at Calcutta, 

 Belgium, Bermuda, and Falmouth, i.e. in places where the 

 ordinary convection currents differ completely both in chaiacter 

 and intensity. 



A comparison of the rainfall with the wind at the end of this 

 paper leads to a conclusion similar to that drawn by Mr. Blan- 

 ford, viz., that rain seldom falls as long as the summer monsoon 

 continues to blow steadily, and Mr. Chambers hence infers, that 

 a strong, damp wind from the seaward, is not the only condition 

 required to produce rain. If this rule is only meant to apply to 

 the place where the wind prevails, it is doubtless correct ; but it 

 seems open to misinterpretation if tal-en in a more general sense, 

 since the laws of cyclonic systems and experience, both tell us 

 that the rea-on why there is little rain on the coast when the 

 sea wind is blowing strongly, is because the area of lowest 

 pressure towards which the wind is spirally blowing is situated 

 in the interior of the country, and that when there is least rain 

 on the coast there is probably most inland. 



Paper X. "Some Results of the Meteorological Observations 

 taken at Allahabad during the Ten Years 1870-79," by S. A. 

 Hill. — This paper, which represents the most complete discus- 

 sion of the climatic elements at a single station in the interior of 

 India that has ever been published, contains much that is vain- 

 able and highly suggestive to the physical meteorologist. To 

 the climatologist it is especially interesting, owing to the inland as 

 well as tropical position of the station. In May and June, 

 Allahabad is one of the hottest places in India, the maximum 

 temperature in the shade often rising above 115° Fahr., while in 

 that terribly hot year, 1S7S, the temperature actually roe up to 

 II9°'S on June 19. 



Nearly all the eh ments are di-cussed by the aid of Bessel's 

 formula, and as it is a paper which cannot readily be reviewed 

 in detail, we propose merely noticing one or two of the most 

 salient conclusions deduced by the author. 



One remarkable feature that comes out from the discussion of the 

 diurnal barometric oscillation, is its "continental" character. Like 

 Yarkand and other typically continental stations, the fall of the 

 night tide is very small, and the ratio of the amplitude of the 

 semi-diurnal to the diurnal component, is not only smaller than 



1 These latter are dealt with in detail in those rapers of Mr. Chambers's 

 which have ak-.ady been alluded t \ 



