December 29, 1910J 



NATURE 



287 



not necessarily a sign of Ctenophore affinities. As in 

 other divisions of the Cceienterata, the determination of 

 true or natural specific distinctions in Medusae from false 

 or accidental differences is extremely difficult. A great 

 many species have been described from the more or less 

 distorted and contracted specimens that are sent to 

 systematic zoologists by the collectors, and it is very prob- 

 able that many of the folds and wrinkles, and even the 

 warts and tubercles, that are relied upon for separating 

 species are due to post-mortem changes. Mr. Mayer deals 

 ith this problem with sound judgment. The work of 

 ;)revious authors is carefully considered and tabulated, so 

 :hat the reader may form his own judgment in each case 

 if he wishes to do so ; but his own opinion, based on a 

 vide experience of living and preserved material, is clearly 

 xpressed. Thus, of the genus Pelagia, no fewer than four- 

 ^n species have been described, of which six are from 

 le .Atlantic Ocean. " .All of the Atlantic species," he 

 5ays, " are closely related one to another, and future 

 researches may demonstrate that they are only geo- 

 graphical races." 



It would be difficult to express adequately our admira- 

 tion of the sevent\-six coloured plates with which this 

 monograph is illustrated. As regards de'icacy of treat- 

 ment and accuracy in detail, they may be regarded as the 

 best series of zoological plates that have been published 

 for many jears. In addition to the plates, there are more 

 than four hundred text illustrations in black and white. 

 The majoritA' of these are copied from the works of other 

 authors, but there are several, such as the two specimens 

 we reproduce, that have not been previously published. 



Important changes in well-known generic names are not 



so common as in some other recent memoirs, but there 



are some which many students of the group will notice 



with regret. Thus the familiar genus Lizzia becomes 



nerged in Rathkea ; Corynitis becomes Linvillea. The 



eneric name Turris, having been used by Humphrey in 



797 for a mollusc, is regarded as preoccupied, and this 



vnus of Medusae becomes Clavula. As examples of 



langes in spelling, we may refer to the genus Irene, 



hich becomes Eirene, and Aurelia, which becomes Aurellia. 



:!ut the most deplorable proposal in this respect is that 



le name Craspedacusta should be used in place of 



-•mnocodium. It is clear from the text that the author 



as made this change with regret, since he realises the 



:--at inconvenience that must be caused by the substitu- 



-n of a name that has been used only once, and in a 



reliminary note, for a name that has been used con- 



stently by all authors, including the writer of the pre- 



rninary note, ever since. That the change has been made 



- due to the mandate of the International Commission on 



/x)logical Nomenclature, who stated that the usage of the 



ame Limnocodium would be "in contravention of the 



; revisions of the Code." No better example could be 



found to show the pressing need of some revision of the 



Code. We cannot close this notice without again express- 



mg our thanks to Mr. Mayer for his most magnificent and 



serviceable memoir. It is really a great work, and will 



mark a great step of progress in the literature of the 



subject. 



MEASURES OF SOLAR PARALLAX.^ 

 T^HE particular value of solar parallax derived from the 

 discussion of any one set of measures is of smaller 

 consequence than the manner in which the result has been 

 achieved. The interest in the problem has shifted. In its 

 present position, the knowledge of the distance of the 

 sun from the earth is less important than the examination 

 and elimination of the causes that affect the accuracy of 

 the measured coordinates obtained from a series of plates. 

 Viewed in this light, Prof. Perrine's paper is of great 

 value, for it puts us in possession of an independent dis- 

 cussion of material that has already been submitted to the 

 most careful scrutiny. 

 We have presented to our examination a numerical 

 stimate of the different constructions that expert know- 



/ "Detenninauon of the Solar Parallax," from Photographs of Eros, made 

 with the Cros-ley Reflector of the Lick Observatory, University of Cali- 

 fornia. By Charles D. Perrine, and others. Pp. v-fo8. (Cameeie 

 Institution of Washington, 1910.) 



NO. 214S, VOL. 85] 



ledge can place upon the same measures. Mr. Hinks, in 

 his elaborate discussion of the solar parallax from photo- 

 graphic observations of Eros, pointed out some discre^ 

 pancies in the Lick results, which he thought required 

 further examination. Among others, he suggested that 

 some of the comparison stars were too distant from the 

 axis of colhmation. Apparently this criticism was justi- 

 fied, and Prof. Perrine has employed in his reductions 'oiiiy 

 those star images which were accurately circular. A. 

 second suggestion, that an error was introduced by the 

 eccentric position of Eros with reference to the stars of 

 comparison, is not accepted. This want of symmetry arose 

 from the plan of choosing the same stars for the morning, 

 and evening observations, a scheme which possesses 

 obvious advantages ; but in a plate taken with Eros 

 always in the centre, the motion of the planet will carry 

 it nearer to, or away from, the more outlying members 

 of the group of stars selected for measurement. The 

 motion of Eros in the interval was about 8'-io', and in a 

 field the available diameter of which is small the distor- 

 tion of the image might outweigh the evident theoretical 

 advantages. 



To test this point Prof. Perrine has made two solu- 

 tions, according to the stars selected, and can find no 

 evidence of systematic error. .Another attempt to explain 

 the observed discrepancy, more of the nature of a sugges- 

 tion than a criticism, was made to depend upon the 

 generally small magnitude of the comparison stars. With 

 a large aperture and the necessity of restricting the field,, 

 there will be a tendency to use fainter stars than in other 

 observatories employing the ordinary photo-refracting, 

 telescope. .As a rule, the stars selected at Lick have been 

 fainter than the planet. Prof. Perrine does not specifically 

 discuss the effect of magnitude, and there is the less 

 necessity, since the value of the solar parallax he obtains 

 does not show any anomalous deviation from the final 

 value adopted by Mr. Hinks. 



The difference of computational results is a point of 

 great interest. The final value of solar parallax derived 

 from the total mass of measures at the command of Mr. 

 Hinks is 8807*, while the same authority obtained from 

 the Lick measures alone 8-815". From the same data 

 Prof. Perrine derives from his own measures 88067'', ^^ 

 identically Mr. Hinks 's result. The problem for solutioh 

 has therefore moved from finding an explanation of the 

 difference of Lick results from the general average to 

 tracing the cause of the disagreement between the Cam- 

 bridge and the Californian computations. The computed 

 probable errors also differ. That attached by Prof. 

 Perrine in his final equation for « is +00025', and by 

 Mr. Hinks +0-0046*. It is a matter for congratulatitMi 

 that such small differences should attract attention and 

 call for explanation. The minuteness of the discrepancy 

 seems to indicate that in modern processes such a degree 

 of refinement has been reached that the disagreement must 

 be attributed to purely arithmetical operations, and has no 

 physical significance. 



AMERICAN VERTEBRATE PALEONTOLOGY. 



nPHE phylogeny of the Felidae forms the subject of an. 

 ■*■ article, by Dr. W. D. Matthew, published in voL 

 xxviii. (pp. 289-316) of the Bulletin of the American 

 Museum of Natural Historj-. -According to the author, 

 the great majority of the extinct members of the family, 

 including all the oldest species, are characterised by a. 

 more or less pronounced development of the upper canines 

 into long, flat-sided tusks. These are the so-called sabre- 

 tooths, or machaerodonts, which date from the Lower 

 Oligocene, typical cats with relatively short upper canines 

 being unknown before the Pliocene. The early sabre- 

 tooths are, however, divisible into two series, one 

 characterised by the extreme length and slenderness of the 

 tusks and the large size of the protecting flange on the 

 lower jaw, and the other by the shorter tusks and smaller 

 flange. Hoplophoneus and Dinictis respectively represent 

 the two series in America. While the derivation of the 

 large Pliocene and Pleistocene sabre-tooths from Hoplo- 

 phoneus has been accepted, the relations of the modern 

 cats to Dinictis have been overlooked. " The evidence 

 appears, however, to indicate that the Dinictis phylum led 



