244 



NATURE 



[Deceuber 2 



1')] I 



of the air currents to our region from some arid con- 

 tinental region like the interior of Asia, which would be 

 very unlikely in summer. 



But the suggestions embodied in Dr. Shaw's communica- 

 tion no longer render necessary such an assumption as 

 small vapour-content of tne air to account for the circum- 

 stances of the drought. For I take it that if a barometric 

 depression were transmitted to the surface strata entirely 

 by variations of pressure occurring beyond the convection 

 region at a height of about 9 kilometres, there would not 

 be engendered the powerful rain-producing convection 

 currents usually associated with cyclonic circulation when 

 a barometric depression takes its origin wholly or partly 

 in causes residing in the lower layers themselves. 



Should this proposition not follow on the lines indicated 

 in Dr. Shaw's letter, I should be glad if he would 

 correct it. L. C. W. Bonacina. 



ilampstead, N.W'., December 2. 



Mr. Bonacina would draw a distinction between baro- 

 metric depression"; transmitted from above and those which 

 take their origin wholly or partly in causes residing in the 

 low(>r layers themselves, and he would attribute a 

 peculiarly rainy character to the latter. I know of no 

 facts which enable us to draw with certainty the distinc- 

 tion in individual cases, and I should have been glad if 

 Mr. Bonacina had cited some examples which were 

 demonstrably of the second type. I cannot speak very 

 positively on the subject. There is a certain hyperbolic 

 region, sometimes shown on weather maps, between a pair 

 of cyclones and alternating anticyclones where I think such 

 depressions may occur. The weather associations of that 

 particular distribution are peculiar ; in the words of a 

 well-known verse : 



When it i^ good, it !•' very very %oodi. 

 But when it is bad it is horrid. 



In summer, thunderstorms with copious rainfall may occur 

 with such a distribution. They are associated with areas 

 of shallow depression — local secondaries — which may be 

 entirely surface-bred ; but the pressure differences involved 

 are so slight that they must be regarded as belonging 

 rather to the embroidery than to the outlines of the 

 pressure distribution of the globe. 



Let me take the opportunity to add that the further study 

 of the physics of rainfall might contribute to the clearing 

 up of some misconceptions about its meteorological 

 significance. Suppose that we regard rainfall as incidental 

 to convection, that is, to the process of righting instability 

 in the lower or middle layers of the atmosphere, which may 

 be caused by the superposition of layers of inverted density. 

 Consider the effect of variations of pressure imposed 

 entirely from above. If the lower layers contain a stratum 

 of floating cloud, it follows from the difference of adiabatic 

 temporature-chinge in cloudy and dry air that compression 

 will produce instability at the lower surface of the cloud 

 and rarefaction will produce instability at the upper 

 surface. Alternate compression and rarefaction, such as 

 would be imposed from above if some giant were playing 

 upon the middle and lower layers of the atmosphere, with 

 their cloud strata, like a concertina, would give alternating 

 instability of different types. With one type rain would be 

 produced. What would happen with the other I have not 

 fully made out ; perhaps the mammato-cumulus cloud may 

 Illustrate it. A good deal depends upon the vertical 

 temperature gradient, of the causes of which at present we 

 know little. 



Certainly if there were no cloud layer to form a locus of 

 instability, compression would simply raise the tempera- 

 ture of the air, and hence its potential dryness. The other 

 stroke of the concertina, the rarefaction, might itself pro- 

 duce cloud, possibly at more than one level, and its con- 

 tinuance would maintain the cloud and the instabilitv 

 necessary for continuous, but fluctuating, rainfall if sup- 

 plies of suitable air were forthcoming. We have here an 

 explanation of cloud layers at different levels and of the 

 well-known tendency towards cloud and rainfall with a 

 falling barometer that is satisfactorily independent of 

 surface conditions. Looking into the details, we can also 

 see that a falling barometer alone is not necessarilv a 

 sufficient cause. 



Parenthetically, may I express the hope that .\" 

 Bonacina or some other student of weather will refer 

 the figures for the dryness of the air of the past sunn: 

 compared with air under similar conditions of win! 



tion, &c., in previous summers? The facts are a<. 



and well worth investigation. There is, 1 have reason 

 think, a high negative correlation between humidity 

 9 a.m. and the duration of sunshine for the day, and t! 

 may point to another factor in the cause of the brillia 

 summer. 



1 hope Mr. Dines will make good his view that the d 

 tribution of temperature in the upper layers is not rclat 

 to the direction of the air currents. When I wrote 1 1. 

 in mind particularly a cold invasion in the upper air 

 July 29, 1908, with a northerly wind, and the appar- 

 increase in intensity of the phenomena of a line squall 

 higher levels. But if we can be assured that cold invasio: 

 begin at the bottom, it will free us from much difficulty. 



In the interesting letter from Dr. Carl Ramsauer pi: 



lished in your issue of December 14 the argument is ba^ 



upon the assumption that during the past summer cc 



i dcnsation in the atmosphere was inhibited for want ' 



i nuclei. The assumption requires proof before it can ! 



{ accepted as a basis of physical reasoning. Supersaturati' 



of air in the free atmosphere is often assumed, but, so 1 



as I know, it has never been demonstrated, and metf< 



ologists, as a rule, do not feel themselves at liberty 



use it. W. \. Snwv 



December 19. 



The contributions in your columns to this subject ;. 

 of great interest. I will venture, with your permission, 

 propound one or two questions suggested by the letter 

 Dr. Carl Ramsauer. On his hypothesis, would it 1 

 follow that all periods of minimum solar activity would ! 

 marked by weather of the 191 1 type? Is this the case? 



On the same assumption, it would seem to follow tl 

 the same type of weather would extend over the wh' 

 earth: has this been the case? We have, I think, hea: 

 of torrential rains in northern Italy in the beginning ' 

 summer, and of the early onset of winter in Canada. 



Again, assuming this hypothesis, not the earth only, \ 

 all the planets would have been affected. Have the si: 

 posed snow-caps of Mars shown any shrinkage? 



Failand, December 15. Edw. I k v 



NO. 2199, VOL. 88] 



The Photography of Ha during Solar Eclipses. 



In Nature for December 7 (p. 182) the report of 

 council of the Royal Society is said to state that at 

 total solar eclipse of April 28 the observers sent out 

 the Joint Permanent Eclipse Expedition secured obser 

 tions with a short-focus prismatic camera showing 

 whole of the hydrogen series, with the additional rema 

 that " this is the first time Ha has been photographic '' 

 an eclipse." 



In the interest of accurate annotation, howevf : 

 statement of the council must be modified, as reference 

 the reports presented to the Royal Society of the eclipse 

 1893 (West Africa) and 1898 (Viziadrug) will show that 

 the Ho line was recorded as a strong line on the photo- 

 graphs obtained with prismatic cameras at both these 

 eclipses by the expeditions from the Solar Physics Obser\"a- 

 tory. It was also recorded bv other observers at the Indian 

 eclipse of 1898. CnARij:s P. Butler. 



The Nematodes of the Thames. 



Referring iO my former communication on this subject, 

 I am happy to say that I have at last discovered the host 

 of at least one of the species of nematode found in the 

 Thames. Mr. Chas. Todd, of Tottenham, has kindly sent 

 me a sample of mud from near the Tower Bridge contain- 

 ing vast numbers of Tubificidae. Among these are Limno- 

 irilus hoffmeisteri, Tubifex campamilatus, and other 

 interesting fresh- or brackish-water worms. The nematode 

 has so far been found only in Tubifex. It gains an 

 entrance into the spermathecae, where it develops to the 

 great inconvenience of its host. Hilderic Friend. 



Swadlincote. 



