i8 



NATURE 



[February 28, 1918 



terms of genera and species. Postulating the original 

 existence of world-ranging generalised family types 

 during an era of uniform conditions, it is argued that 

 the differentiation of these primitive types was in re- 

 sponse to the progressive differentiation of their con- 

 ditions. The distribution of families is treated Statis- 

 tically, and it is shown that whilst they largely ignore 

 the cleavage of the land into two great masses, diverg- 

 ing from the north, they respond in marked degree to 

 the differentiation of the climatic zones. The paper 

 •ends with the statistical treatment of the larger groups 

 l:)ehind the families, and it is shown that whilst the 

 Dicotyledons display a much greater tendency to de- 

 tachment from the tropics than do the Monocotyledons, 

 the Sympetalag stand foremost in this respect amongst 

 all the primary groups. 



Physical Society, February 8. —Prof . C. V. Boys, 

 president, in the chair. — Prof. C. V. Boys : A recording 

 thermometer. This instrument was designed and con- 

 structed to go into the case of a- regulator clock. The 

 thermometric element consists of a rod of ebonite 

 within a glass tube. The differential expansion is 

 determined by a pair of levers giving a movement of 

 I in. for io° F. The drum carries an ordinary baro- 

 meter chart, and is driven at' such a speed that a two- 

 hour interval of | in. is passed in twenty-four hours. 

 The drum is driven by friction by means of a cord 

 from b^low the driving weight of the clock by an e'** 

 •arrangement, in virtue of which when the clockweight 

 descends the drum turns, but when the clock is wound 

 the drum remains at rest. The instrument is designed 

 with a view to easy construction and accuracy. It is 

 ■extremely rigid, and much more magnification might 

 be used. An alternative design on the same lines to 

 go into a recording barograph is also given. — S. D. 

 Chalmers : The primary monochromatic aberrations of 

 a centred optical system. The paper describes approxi- 

 mate methods of treatment of the first-order aberra- 

 tions of a centred optical system. Two methods are 

 used, one primarily suited to the case where the separa- 

 tion of the surface is small, and the other more suited 

 for use where the separations vitally affect the design. 

 The aberrational defects are expressed as lateral aber- 

 rations — i.e. as defects measured in the focal plane of 

 the system. The procedure adopted is to express the 

 aberrational defect of a single surface in terms of the 

 constants of the surface, and the perpendicular distance 

 of the ray considered from the centre of curvature of 

 the surface. The value of this perpendicular can be 

 expressed in terms of the co-ordinates of the ray in 

 any chosen medium, and thus the aberration due to 

 ■each surface can be expressed in terms of the co- 

 ordinates of the chosen ray, in such a way that the 

 aberrations of the individual surfaces can be summed. 



Optical Society, February 14.— Prof. F. J. Cheshire, 

 president, in the chair. — T. . Y. Baker: Reflecting 

 prisms. The author advocated the use of prisms in 

 place of mirrors as being easier to mount and as re- 

 ■quiring, in many cases, no silvering. A series of suit- 

 able designs for certain double reflecting prisms, in 

 which the light is made to deviate by a fixed amount, 

 was shown ifor angles 0°, 15°, 30°.. .90°, the form in 

 ■each case being that which gave maximum aperture to 

 the prism. A special form of triple reflecting prism, 

 with angles of 30°, 30°, and 120°, was described, which 

 the author considered would be much more suitable for 

 the horizon glass of a sextant than the customary plane 

 mirror, as it would enable the telescope to be placed 

 close up to the prism without any danger of cutting off 

 light between the two mirrors. To avoid this happen- 

 ing in an ordinarv sextant, the horizon glass has to be 

 set at an angle of 75° to the axis of the telescope, and 

 the latter set well back, so that the angular aperture 

 NO. 2522, VOL. lOO"! 



is equivalent to 15°, whereas with the prism described 

 the aperture would be about 50°, and angles could be 

 measured up to 170°, in place of the maximum range 

 of 150° in an ordinary sextant. 



Royal Meteorological Society, February 20.— Sir Napier 

 Shaw, president, in the chair.~F. A. Bellamy : The 

 barometer record at the Radcliffe Observatory, Oxford, 

 with special reference to Prof. Turner's suggested dis- 

 t:ontinilities. Prof. Turner has in several papers to the 

 Royal Meteorological Society claimed that meteoro- 

 logical history is divisible into "chapters" of an 

 average length of six and a half years each, and has 

 assigned the -dates at which a new "chapter" opens 

 (when there are abrupt discontinuities in meteoro- 

 logical phenomena) with considerable precision for the 

 last two centuries. The evidence has hitherto been 

 based upon the monthly mean values of rainfall and 

 temperature, and he has shown that these monthly 

 means differ systematically in the " even " chapters as 

 compared with the "odd." To examine whether such 

 systematic differences extended to periods shorter than 

 one month, Mr. Bellamy has analysed a series of sixty- 

 two years' daily barometer records made at the Rad- 

 cliffe Observatory, Oxford, many of them made by 

 himself while there in 1891-92. He concludes from 

 his analysis that for periods from one to six months there 

 is a decided difference in the even and odd chapters for 

 atmospheric pressure, even for one station, as there 

 has been shown for rainfall arid temperature from 

 many stations in the world, and that the matter of 

 discontinuities is supported. — Dr. C. Chree : The 

 diurnal variation of barometric pressure at seven 

 British observatories, 1871-82. The diurnal variation 

 of atmospheric pressure, as of any other element, can 

 be analysed in Fourier terms or " waves " of periods 

 twenty-four, twelve, and eight hours. In the case 

 of the mean diurnal variation from all months of the 

 year combined, the twelve-hour wave appears of a 

 very dominant character, the amplitude being nearly 

 the same for all stations in the same latitude, and the 

 phase referred to local mean time being everywhere 

 nearly the same. The twenty-four-hour wave, on the 

 other hand, is very variable, and the eight-hour wave 

 small compared with the twelve-hour wave. Taking 

 the seven stations, Valencia, Armagh, Glasgow, Aber- 

 deen, Stonyhurst, Falmouth, and Kew, it is pointed 

 out that the comparative unimportance of the eight- 

 hour wave in the mean diurnal inequality for the year 

 arises in great measure from the large difference that 

 presents itself between the phase angles in two different 

 seasons of the year. In individual months the eight- 

 hour wave, though considerably smaller than the 

 twelve-hour wave, is far from negligible, and the 

 phenomena presented by the eight-hour wave at the 

 seven British stations exhibit almost as close a simi- 

 larity as those presented by the twelve-hour wave. 

 The paper also considers a theory as to the nature of 

 the twelve-hour wave recently advanced by Dr. G. C. 

 Simpson. 



Manchester. 



Literary and Philosophical Society, February iq. — Mr. 

 W. Thomson, president, in the chair. — Dr. J. Stua; 

 Thomson : The occurrence of Cavernidaria LiUkcnii. 

 Koll, in the seas of Natal.— L. Stanley Jast : The 

 necessity for a technical library for Manchester and 

 district. Manchester should lead in that provision of 

 technical libraries which must form a not inconsidc 

 able part of our equipment for shouldering our il. 

 share of the commerce of the world after the war. 



I Edinburgh. 



Royal Society, February 4. — Dr. John Home in the 



chair. — Dr. J. Stuart Thomson : The morphology of the 



I Prosencephalon of Spinax as a type of Elasmobranch 



