3i8 



NATURE 



[November 3, 192 1 



on botany. Some of them are unique, the plates 

 being copies of the original issues, executed by R. 

 Simk6, an Hungarian painter, while the text is neatly 

 typewritten. This is the case with Sibthorp's "Flora 

 Graeca," the 966 plates of which have been redrawn, 

 apparently with extraordinary success, for they are de- 

 scribed as being more artistic than those of the 

 original issue, the merits of which are wel.l known. 

 Several of Jacquin's rare works are included in the 

 collection, amongst them being his " Hortus Botani- 

 cus Vindobonensis," "Flora Austriaca," and " I cones 

 Plantarum Rariorum." There is also a complete set 

 of W. Griffith's "Posthumous Papers," Waldstein and 

 Kitaibel's " Descriptiones et Icones Plantarum Rario- 

 rum Hungariae," Reichenbach's • " Iconographia 

 Botanica, " and "Icones Florae Germanicae et Hel- 

 veticae," Lapeyrouse's "Figures de la Flore des Pyre- 

 nees," and Weinmann's " Phytanthozaiconographia." 

 In Science Progress (No. 62) Mr. F. W. Flattely, 

 in an article on "Some Biological Effects of the 

 Tides," discusses the consequences which tides and 

 tidal action have had on the life of the sea-shores. 

 The abundance of sedentary or fixed forms and the 

 widespread occurrence of the phenomenon of stereo- 

 taxis among free-swimming species are traced to the 

 direct action of wave impact. Reference to such 

 semi-marine animals as Ligia, certain species of Lit- 

 torina, Birgus, and Periophthalmus, is made to show 

 that the population of the land from the sea took 

 place via the shore as well as by way of the rivers, 

 and it is suggested that the daily and monthly changes 

 in tidal level, by producing alternating aquatic and 

 terrestrial conditions, made the shore zone an effec- 

 tive bridge between the land and the sea, and pro- 

 vided the opportunity for marine animals to. attempt 

 the conquest of the land. The case of Convoluta is 

 quoted to show that the daily rhythm of the tides 

 has imposed a periodicity upon the behaviour of 

 shore forms which tends to become impressed on the 

 animals and to persist after their removal from tidal 

 influence. The far-reaching consequences of such an 

 effect on animal behaviour are suggested rather than 

 insisted on. 



In a paper in the Philosophical Magazine for Septem- 

 ber Drs. Dorothy Wrinch and Harold Jeffreys discuss 

 some of the fundamental principles of scientific infer- 

 ence. They assume that scientific arguments must 

 conform to the rules of pure logic— an assumption 

 which, of course, is by no means universally admitted. 

 They then inquire what conditions must be fulfilled 

 in order that a proposition shall have a finite prob- 

 ability of truth as the result of empirical verification, 

 the term "probabiUty " being used in a sense which 

 they have expounded in a previous paper (Phil. Mag., 

 December, 19 19). They conclude that if all possible 

 forms of the law are equally probable a priori, then 

 no amount of empirical verification can establish a 

 finite probability in favour of one law rather than 

 another. In order that a finite probability may be 

 established it is necessary that the class of possible 

 laws should form a well-ordered series in which each 

 term is more probable a priori than its predecessor, 

 NO. 2714, VOL. 108] 



and that the probabilities of this class form a con- 

 vergent series. The supposition that this necessary 

 condition for valid scientific inference is fulfilled is 

 equivalent to the admission of some principle of "sim- 

 plicity." Thus, like most of those who discuss the 

 foundations of science, they arrive at very familiar 

 conclusions as a result of very complicated arguments, 

 but that result does not detract from the value or the 

 interest of their inquiries. 



The Greenwich observations for the twelve months 

 from October, 1920, to September, 1921, give 

 1243 in. as the total rainfall, which is only 51 per 

 cent, of the normal for the last hundred years. This 

 is the smallest amount on record for the correspond- 

 ing period, the next smallest being apparently from 

 October, 1897, to September, 1898, when the fall was 

 1475 in., or 60 per cent, of the normal. For deficiency 

 this was followed bv 1674 in. from October, 1863, to 

 September, 1864, which is 69 per cent, of the hundred- 

 year normal. The rainfall has been below the average 

 in each of the last thirteen months. This is the 

 longest dry period in the last hundred years with 

 the single exception of fifteen months from Novem- 

 ber, 1846, to January, 1848, during which time the 

 rainfall for the twelve months from November, 1846, 

 to October, 1847, registered only 16-26 in. October is 

 usually the wettest month of the year, but this year 

 it had a deficiency of more than i^ in. Different 

 weather conditions prevailed in the early and latter 

 halves of October ; the first half of the month ex- 

 perienced summer weather, while in the latter 

 half the weather was cooler and less sunny. Each 

 day from October i to 15 was abnormally warm ; the 

 mean dailv excess over the normal was 10-7° F., the 

 greatest excess being 17° F. on October 6, and there 

 were eight days with an excess of 137° F. or more. 

 The duration of sunshine at Greenwich from 

 October i to 15 was loi hours, which is 4 hours 

 more than the normal for the whole month. 



The Department of Commerce, Bureau of Stan- 

 dards, Washington, has issued Scientific Paper No. 

 416, entitled, "Preparation of Galactose." Owing to 

 the demand made by bacteriologists for galactose and 

 its derivatives, a convenient method is described for 

 preparing this compound from lactose. One kilogram 

 of lactose is hydrolysed by boiling for two hours with 

 2-5 litres of water and 50 grams of sulphuric acid. 

 The solution is neutralised with barium carbonate, 

 filtered, and concentrated. The galactose is 

 crystallised from the resulting syrup by the addition 

 of a mixture of one part of ethyl and two parts of 

 methyl alcohol. The yield of crude sugar is about 

 27 per cent, of the lactose taken. The galactose is 

 purified by concentrating to 75 per cent, of total 

 solids, under diminished pressures, a 25 per cent, 

 solution to which a little glacial acetic acid is 

 added. The material is warmed to 60-70°, trans- 

 ferred to a beaker, and 95 per cent, alcohol added 

 to saturation. After standing overnight the crystals 

 are filtered, washed, and dried. 



The presidential address delivered by Mr. G. W. 

 Watson on October 12 last, at the Institution of 



