28 



NA TURE 



[Mav lo, 1894 



OUR BOOK SHELF. 

 The Natural History of Plants, from the German of 



Prof. Anton Kemer von Marilaun. By Prof. F. W. 



Oliver, M.A., D.Sc. (London, Glasgow, and Dublin : 



Blackie and Son, 1S94.) 

 The high praise we gave to Prof Kerner's Pflansen- 

 leben when it appeared, makes it almost unnecessary for 

 us to say much about the English edition now in course 

 of publication, and which will be completed in sixteen 

 monthly parts. The German work was said in these 

 columns to be " the best account of the vegetable king- 

 dom for general readers which has yet been produced." 

 This judgment can also be applied to the translation 

 which Prof. Oliver has made, with the assistance of Miss 

 Marian Busk and Miss Mary Ewart. In translating a 

 work, some of the brilliancy of the original is neces- 

 sarily lost. It is difficult, however, to find awkward 

 expressions in the pages before us ; in fact, very few of 

 the idiomatic phrases of the original work have survived. 

 And the translation is scientifically accurate, as well as 

 entertaining and instructive. Lovers of nature will find 

 ever)' page of the book interesting, and the serious 

 student of botany will derive great advantage from its 

 perusal. The illustrations are beautiful, and, what is ; 

 more necessarj-, true to nature. The complete work 

 contains about one thousand engravings on wood, and 

 sixteen plates in colours. Botanical science will benefit 

 by the issue of Prof. Oliver's edition of a splendid book. 



Notes on some of the more Common Diseases in Queens- 

 land in relation lo AtmospJieric Conditions, 18S7-91. 

 By David Hardie, M.D., Hon. Physician Hospital 

 for Sick Children, Lady Bowen Maternity Hospital, 

 Brisbane. (Brisbane: Beal, 1S93.) 



The author of this work has a most important aim 

 in view, viz. to establish the connection between the 

 weather and the prevalent diseases in Queensland, 

 and expresses a hope that, in time, he will be able, if 

 furnished with a forecast of the weather, to predict with 

 certainty the diseases likely to predominate during the 

 various seasons of the year, and thus to lay the 

 foundations of a practical system of preventive 

 medicine. 



The conclusions are so interesting that some of his 

 leading results may be briefly given. 



The annual death-rate of Queensland per 1000 popu- 

 lation is 151 1, varying from 1338 in August and Sep- 

 tember (spring) to i6'28 in November to March (summer 

 and early autumn), as contrasted with the death-rate of 

 Great Britain for the year, 188. 



The yearly mortality is lowest in West Southern 

 Queensland (Darling Downs and Warrego), where it is 

 only 892 per 1000, a little over half of the average 

 Queensland rate, and the highest is 3470 in West 

 Northern Queensland (Normanton), a tropical region at 

 the extreme north of the colony. This would point to 

 the great advantage of altitude, combined with dryness, 

 as seen on the Darling Downs, over marine influence 

 and moisture to be found in the sea-coast districts. 

 When we come to different classes of disease, we find 

 that diphtheria, though specially prevalent in April, May, 

 June, and July (winter), is endemic to some extent in all 

 seasons, and causes a mortality of 21 5 per cent. What- 

 ever contributes towards cold and dampness of the air 

 during the autumn and winter, causes an increase in the 

 death rate (rom diphtheria, and, according lo Dr. Hardie, 

 the neighbourhood of swamps and marshes have con- 

 siderable influence in this respect. Whooping-cough, 

 on the other hand, attains its maximum during the warm 

 and moist months of the year, and its close connection 

 with a medium temperature for all seasons of the year 

 with high relitive humidity, is considered to support the 

 assumption of its germ origin. 



NO. I2bo, VOL. 50] 



Phthisis is common along the eastern portion of the 

 colony from Cooktown to Brisbane, reaching a maxi- 

 mum of 12S6 per cent, in the Rockhampton district; 

 but this high mortality is partly to be attributed to the 

 large Polynesian population, employed on sugar planta- 

 tions in these districts, who are specially liable to phthisis. 

 In the western and northern districts, however, it is much 

 less prevalent, and the average percentage of deaths 

 from all causes gives to phthisis S 75. The months when 

 the mortality is greatest are July, .August, and September, 

 and there seems to be no special connection between 

 atmospheric moisture and phthisis mortality, but a low 

 temperature in summer and autumn is favourable to a 

 low phthisis mortality. Acute respiratory diseases, such 

 as pneumonia, pleurisy, and bronchitis, are observed all 

 over the colony, and vary in mortality in different parts ; 

 the highest on the coast and the lowest inland, the 

 months of highest mortalitybeing June, July, August, and 

 September ; the maximum is reached during and imme- 

 diately after the colder period of the year. 



The book, with its copious and valuable tables, is an 

 honest attempt to deal with a very difficult problem, and 

 thoroughly merits success, and if the author will only 

 persevere in his researches, still more important results 

 may follow. 



LETTERS TO THE EDITOR. 



\Thc Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondent:. Neither can he undertake 

 to return, or lo correspond with the •.vrilers of, rejected 

 manuscripts intended for litis or any other pari d/Nature. 

 No notice is taken of anonymous communicalions.\ 



Panmixia. 



I AM much obliged to Prof. Weldon for having so promptly 

 answered my request, and hope that his example will be 

 lollowcd by any naturalists who may have any olher grounds 

 for questioning the doctrine of Panmixia. Meanwhile, how- 

 ever, there are two or three points touching which I should 

 like to he sure that we correctly underslami e.ich other. 



(1) Iliihtrtoall naturalists who have written upon the subject 

 have agreed, that " Ihe survival- mean musi (on cessalioit ,<' 

 selection) fall lo Ihe birth-mean." And, in now quesli.)ning this 

 view, Piot. Weldon appears to contemplate the diflerence 

 between birth- and survival-means of only the first generation, 

 which would be very unfair. Again, 1 do not follow I'rol 

 Weldon's meaning in what he says with regard to another ])uint. 

 .\ssuredly "every statisticLin knows" that selection can main- 

 tain the " mean height of a regiment " at ojh inches, by 

 enrolling only those individuaN who are either " more than 66 

 and less than 69 inches high." But this wmld be arliftcial 

 (i.<. intentional) selection. The " cases " to which he alludes, 

 where natural selection could destroy individuals nearest the 

 mean line, while favouring those which lie at greater distances 

 both aliovc and belo:v this line, must be very exce|jti.>nal. 



(2) As regards the second cause of degeneration under 

 Panmixia (vi?.. atavism), Prot. Weldon says merely that it is 

 "not demonstrated by any statistics." This is true enough. 

 Hut the same has to be said of natural selection. Whether in 

 ihe building up of a structure by n.itural selection, or in the 

 subsequent hreaking down of a structure by atavism on the 

 withdrawal ol seleciion, the statistical method is eiiu.illy unavail- 

 able for lesiing cither theory : in both cases the most effective 

 variations (i.e. deviations from the mean) at any given time arc 

 those which arc most numerous, and therclore most muiulc. 

 Hence, in both c.ises the best " dem .nslration " of the theory 

 which can be offered is that which is yielded by the parallel 

 facts in our domesticated animals. 



(3) The only objection which is urged by Prof. Weldoft 

 against the last of the three causes which I mentioned (i.e. 

 irrcgulaiitics of heredity when unontrollc I by selection) is one 

 which tells against the theory of Panmixia only because it does 

 so against that ol Natural Selection. As I undcrsiand, the 

 argument is, "Natural selection is in most cases an imperfect 

 agent in the adjustment of organisms " : er^o, the cessation of 



