May 30, 1895] 



NA TURE 



99 



short intervals of time. But this great certainty is based 

 upon experience and observation, rather than upon purely 

 thermo-dynamic principles. 



The evidence of decisive progress in forecasting is 

 wanting. Nor does the author hold out a very sanguine 

 hope of the possibility of issuing in the immediate future 

 successful weather forecasts over large districts from a 

 central bureau. There are not more than six to 

 twelve occasions, in the course of a year, for any part of the 

 countr)-, he tells us in the preface, "where successful pre- 

 dictions can be made, and for some places successful 

 predictions are never possible." " Successful continuous 

 predictions for every day are not possible." This is 

 the opinion of one who apparently has ample means of 

 forming an adequate judgment. It is the outcome in a 

 country where the opportunities of framing forecasts are 

 many and favourable. The service is well supplied both 

 with funds and officers, the vast telegraphic system of 

 the country is at the disposal of the Weather Bureau, 

 the area over which the data are collected is extensive 

 enough to enable the whole development of a storm to 

 be watched and reported, while the favourable situation of 

 Washington, in the extreme east of the continent, is a 

 point not to be omitted. Yet after years of trial, the 

 opinion of one who apparently has official connection with 

 the system, or is at least well supplied with information 

 from the Bureau, is, that the complete solution of the 

 problem is not only impossible, but is only practically 

 effective on the average less than once a month. If this 

 be the result under favourable conditions, what, it may 

 be asked, is the system worth in England, where our 

 insular position cuts off the supply of any information 

 from the West, the direction in which our principal storms 

 approach, and the intelligence from the East has to be 

 supplied by the courtesy of many nationalities, and more 

 or less hampered by different telegraphic systems. 



To return to the book, howe\ er, which in some respects 

 is a little disappointing. There is an occasional appear- 

 ance of hurry in the compilation of the work, which has 

 sometimes prevented the author expressing himself with 

 sufficient clearness, and with the reservations which are 

 sometimes necessary. For instance, we are told, on p. 3, 

 that there is less oxygen in the air when the wind is from 

 the south, than when the direction is north. This 

 may be true for the district in which the author lives, but 

 as there is no indication where this particular locality is 

 situated, and the preface is not even dated, we are left 

 to infer that the remark applies to the earth generally, 

 which can scarcely be correct. Again, on p. 184, in the 

 description of secondary low pressures, occurs this 

 sentence. " In Fig. 29, thunderstorms are very apt to 

 occur with secondary low pressures.'' This statement is 

 certainly a puzzle. On p. 190 we are referred to a map 

 on the adjoining page. There is no map there, although 

 this map is referred to in the list of illustraticms. 

 Headers w ill, however, find it at the end of the book. 

 Sometimes, too, facts which are easily verified are not 

 quoted with accuracy. On p. 5, the dates of the earth's 

 perihelion and aphelion passage are given as December 

 23 and June 21 respectively. The area of the Caspian 

 Sea is given on p. loi as over 200,000 square miles, and 

 on p. 201 as 180,000 square miles. But these and 

 KO- 1335. VOL. 52] 



many other small blemishes can be removed in a future 

 edition. 



We are more concerned to look at the work as a whole, 

 and to consider what special service is it likely to render 

 among the host of meteorological treatises that are con- 

 tinually appearing on one or other side of the Atlantic 

 We have, of course, the ordinary chapter on meteorological 

 instruments ; we have the cloud classification ; we have 

 the description of the rain and hail and snow, that too 

 frequently make life unpleasant ; together with all the 

 winds that blow, or are likely to blow. And the oft-told talc, 

 it must be confessed, is repeated in rather a jerky manner, 

 partaking of something of the manner of a dictionar), 

 wherein one is treated to a collection of definitions. 

 The last chapters of the book are undoubtedly the 

 best. There the author has something to tell us of 

 processes not generally described in books like the 

 present. To the charm of novelty is added the ad- 

 vantage that we feel we are listening to a practical 

 expert, who can tell us all that is worth knowing about 

 river-floods and overflows. 



We come now to the second volume under notice. 

 Fortunately in this country we are not frequently 

 troubled by the overflow of rivers and the consequent de- 

 struction of property on the banks, and therefore the 

 subject with us receives scant attention. Probably for 

 this reason the report of the Meteorological Council is 

 silent on such matters, though at times like last autumn, 

 the inhabitants of Eton, Oxford, and the Thames Valley 

 would have been gratified by a timely warning. It may 

 have been that warnings were given, but from the absence 

 from the Report of any mention of machinery adapted to 

 this end, one would infer that this is an inquiry the 

 Council do not consider worthy of their attention. Far 

 different is it with the Astronomer at Sydney, whose latest 

 report is mentioned at the head of this notice. There the 

 subject is forced on the attention of scientific men : and 

 on the unscientific, too, if he happen to live in a district 

 where, as Mr. Russell reports, the rise of a river was so 

 rapid that in less than two hours a part of a town was 

 covered to a depth of three or four feet, and the people 

 were glad to escape with their lives at the sacrifice of their 

 property. Mr. Russell has great difficulties to contend 

 with. He has not only the small equipment peculiar to a 

 comparatively new colony, imperfect data, and the slow 

 accumulation of facts, but the first warning of the rise of a 

 flood may occur in uninhabited or thinly populated dis- 

 tricts, with which communication is slow and uncertain. 

 The American Bureau has not to struggle against these 

 disadvantages, but the problem depends upon so many 

 variable quantities that the complete solution is practically 

 impossible. 



The author of the treatise on meteorology lays 

 it down that very little connection can be traced between 

 meteorological laws and river floods, except perhaps in 

 cases where the quantity of water is dependent upon the 

 melting of the snow. In temperate zones, floods occur 

 without any very noticeable great rainfalls. Intermittent 

 rain may cause a river to rise very slowly, and almost im- 

 perceptibly, till it be bank-full, when a moderate rain 

 makes the river overflow. Neither is there any decided 

 connection between the river slopes and velocity, so that 



