190 



NATURE 



[December 22, 1892 



end of which forms a large prominence above and in front of the 

 orbit. The occiput has only one (the lower) post-temporal fossa 

 open on each side. The maxilla is produced into a tooth-like 

 prominence, which occupies a similar position to the tusks of 

 Gordonia; but the bone is too thin to have supported a tooth, 

 and in all probability it was covered by a horny beak. The 

 lower jaw has a strong symphysis, a distinct lateral vacuity, and 

 the oral margin, at the foot of each ramus, bears a rugose 

 prominence. 



Elginia niirabilis is the name proposed for the skull of a 

 reptile, which, on account of the extreme development of horns 

 and spines, reminds one of the living lizards Moloch and PAry- 

 nosonia. The exterior of this skull is covered in by bony plates, 

 the only apertures being the pair of nostrils, the orbits, and the 

 pineal fossa. The surfaces of the bones are deeply pitted, as in 

 crocodiles and labyrinthodonts. The horns and spines, which 

 vary from \ in. to nearly 3 in. in length, are found upon nearly 

 every bone of the exterior. The development of the epiotics, 

 and the arrangement of the external bones, resemble more the 

 Labyrinthodont than the reptilian type of structure ; while the 

 palate, on the other hand, conforms more nearly to the Lacer- 

 tilian type, and, with the exception that the pteryr^oids are 

 united in front of the pterygoid vacuity, agrees with tne palate 

 of Iguan • and Sphenodon. There are tour longitudinal ridges 

 along the palate, some of which seem to have carried teeth. The 

 oral margin was armed with a pleurodont dentition, there being 

 on each side about twelve teeth with spatulate crowns, laterally 

 compressed and serrated. With the exception of the smaller 

 number of the teeth, we have here, on a large scale, a repetition 

 of the dentition of Iguana. This peculiar skull seems to show 

 affinities with both Labyrinthodonts and Lacertilians, and is 

 unlike any living or fossil form ; its nearest, though distant, ally 

 'apparently being the Pareiasaurus from the Karoo beds of 

 South Africa. 



Linnean Society, December i.— Prof. Stewart, President, 

 in the chair. — A letter was read from the Rev. Leonard Blome- 

 field, expressing his high appreciation of the compliment paid 

 him by the presentation of the illuminated address which had 

 been signed by the Fellows present at the last meeting of the 

 Society and forwarded to him. — Messrs. H. and J. Groves ex- 

 hibited specimens of several Irish Characea collected during the 

 past summer. Nitella tenussima from Westmeath and Galway 

 had not been previously recorded from Ireland, and a large 

 form of N. gracilis from two lakes in Wicklow had been only 

 once previously met with. Referring to the former, Mr. H. 

 Groves remarked that although it might be expected to occur in 

 all the peat districts it had only been found in two widely 

 separated localities in England, namely, in the Cambridgeshire 

 Fens and in Anglesea. — Mr. A. Lister made some remarks on 

 the nuclei of Mycetozoa, exhibiting some preparations under 

 the microscope. — Mr. E. Cambridge Phillips forwarded for ex- 

 hibition a hybrid between red and black grouse, which had 

 been shot in August near Brecon. — Mr. J. E. Harting exhibited 

 and made remarks on some coleopterous larvae which had been 

 vomited by a child at Tintern, and had been forwarded by the 

 medical attendant, Dr. J. Taylor Brown, for identification. The 

 precise species had not been determined, but was considered 

 to be allied to Blaps mortisaga, Mr. Harting drew attention 

 to the fact that cases of voiding coleopterous larvae were men- 

 tioned by Kirby and Spence (7th ed. p. 71), and by the late 

 Dr. Spencer Cobbold in his work on parasites (1879, p. 269).— 

 Mr. D. Morris exhibited some tubers of Calathia allonia, eaten 

 as potatoes in Trinidad, where it is known as Tapee Nambour. 

 Apparently a corruption from the French topinambotir (arti- 

 choke). — A communication was read from Mr. j. H. Hart, of the 

 Botanic Gardens, Trinidad, on CEcodotna cephalotes and the 

 fungi it cultivates.^ — Prof. F. Jefifry Bell contributed a short 

 paper on a small collection of Crinoids from the Sahul Bank, 

 North Australia, some of which were new, and Mr. Edgar 

 Smith communicated descriptions of some new land shells from 

 Borneo. — The meeting adjourned to December 15. 



Physical Society, December 9. — Mr. Walter Baily, Vice- 

 President, in the chair. — The Chairman announced that an 

 extra meeting would be held on January 13, 1893. — Prof. S. P. 

 Thompson's communication on Japanese magic mirrors was 

 postponed. — Mr. W. B. Croft read a paper on the spectra of 

 various orders of colours in Newton's scale. After referring to 

 the definition of the oi-der of colours by reference to the retarda- 

 tion in wave-lengths, produced by different thicknesses of selenite 



NO. 1208, VOL. 47] 



between crossed polarizer and analyzer, the author went on to 

 say that several books on optics implied that the number of 

 bands in the spectra of these colours was the same as the order 

 of the colour. On obtaining selenites of the first four orders 

 of red from Messrs. Steeg and Renter, he found that the first 

 three orders gave one dark band each, and that of the fourth 

 order three dark bands. Further experiments showed that the 

 thicknesses of the selenites were in the proper proportions re- 

 quired to give the first four orders of red. The numbers of 

 bands, the author explained, depended on the numerical possi- 

 bilities of wave-length within the visible spectrum — that is, 

 whether a multiple of the wave-length of one visible wave can 

 be another multiple of a different wave. For example, taking 

 the visible spectrum as extending from A (0*000760) to H 

 (0-000394) and the wave-length of the line E in the green as 



000527, it was shown that the first order of red was due to 

 extinction of green by a thickness of crystal proportional to 



1 X 0-000527, and would give one band in the green. For the 

 second order, the thickness of crystal was proportional to 



2 X o'ooo527, viz. o •001054, and this number was no other 

 integral multiple of any other wave-lengih between A and H ; 

 consequently there could only be one band. Similarly it was 

 shown that the third order of red could only have one band or 

 possibly produce a shortening of the spectrum. With the fourth 

 order of red three bands were obtainable, for 4 x 0*000527 = 



3 X 0-000703 and = 5 X 0-000422. Three bands were there- 

 fore possible near E, A, G, respectively. At the conclusion of 

 his paper, Mr. Croft directed attention to a very simple form of 

 diffraction apparatus, by which most of the ordinary diffraction 

 phenomena could be well seen, and which also served for 

 spectrum observations. Mr. H. Miers pointed out that in 

 Lewis Wright's " Practical Optics " a chart showing the bands 

 corresponding to the first four orders of red was given. So 

 far as he was aware, the subject was not fully discussed in the 

 book. In reply, Mr. Croft said he had noticed Mr. Wright's 

 chart, but believed the text implied that the number of bands 

 should be the same as the order of the colour. Tyndall made 

 definite statements to that effect.— Dr. W. E. Sumpner read a 

 paper on the diffusion of light. The influence of diffusion in 

 increasing the illumination of rooms and open spaces, had not, 

 in the author's opinion, been sufficiently appreciated. Being 

 impressed with the great importance of the subject, he was led 

 to make determinations of the co-efficients of reflection, absorp- 

 tion, and transmission of diffusing surfaces. To give precision 

 to terms sometimes vaguely used, several definitions were pro- 

 posed. Reflecting power was defined as the ratio of the amount 

 of light reflected from a surface to the total amount of light 

 incident upon it ; illumination of a surface, as the amount of 

 incident light per unit of surface ; unit quantity of light as the 

 flux of radiation across unit area of a sphere of unit radius at 

 whose centre a unit light is placed ; and brightness as the candle- 

 power per unit area in the direction normal to the surface. 

 Denoting these quantities by 17, I, Q and B respectively, arid 

 assuming the cosine law of diffusion {i.e. the candle-power in 

 any direction is proportional to the cosine of the angle between 

 the direction and the normal to the surface) it was shown that 

 TT B = ■»? I, and that the average illumination (!') of the walls of 

 a room is related to the illumination (I) due to the direct action 



of the lights as expressed by the formula I' = - If the re- 

 flecting power of the walls, &c., be 50 per cent., i\ = J, and 

 r = 2I, whilst if Tj = 0-8, a number approximately true for while sur- 

 faces, then I' =51. The illumination due to the u alls may, there- 

 fore, be far more important than that due to the direct rays from 

 the lights. When the surfaces consist of portions of different 

 reflecting power, the average reflecting power may be found 



from the equation ti 



_7JlAi -f TJ.iAj 



&c. , A being the total 



surface, and Aj, Ag, &c., the areas of surfaces whose reflecting 

 powers are 171, r/j, &c., respectively. This law is shown to be 

 quite accurate for spherical enclosures. In mensuring reflecting 

 power, the surface was attached to a large screen of black velvet 

 placed perpendicular to a 3-mefre photometer bench. Two 

 lights were used, one a Methven 2-candle standard placed at 

 the end of the bench remote from the reflecting surface, and the 

 other, a glow lamp of about 20 candle power, was attached 

 to a slider which also carried a Lummer-Brodhun jihotometer. 

 The glow-lamp served to illuminate the reflecting surface, but 

 the photometer was screened from its direct rays. The formulae 

 used in reducing the observations are worked out in the paper 



