i;8 



SCIENCE. 



natives of that island "barbarous people" in the same 

 sentence in which he tells us of their kindness and hos- ; 

 pitality. This simple and purely negative meaning of | 

 the word barbarian has been lost to us, and 

 it has become inseparably associated with char- 

 acteristics which are indeed common among uncivilized 

 nations, but are by no means confined to them. The 

 epithet "savage," of course, still more distioctly 

 means something quite different from rude, or 

 primitive or uncultivated. The element of cruelty 

 or of ferocity is invaiiably present to the mind where we 

 speak of savagery, although there are some races — as 

 for example, tne Eskimo — who are totally uncivilized, ! 

 but who, in this sense, are by no means savage. 



And this may well remind us that, as we have found 

 it necessary to define to ourselves the condition which we 

 are to understand by the word civilization, so it is not 

 less essential to define and limit the times to which we 

 are to apply the word primeval. For this word also is 

 habitually used with even greater laxity of meaning. It 

 is often employed as synonymous with primitive, ard 

 this again is applied not only to all times which are pre- 

 historic, but all conditions even in our own age which are 

 rude or savage. There is an assumption that, the farther 

 we go back in time, there was not only less and less 

 extensive knowledge of the useful aits, — not only simpler 

 and simpler systems of life and polity. — but also that 

 there were de-prr and deeper depths. of the special char- 

 acteristics of the modern savage. We have, however, 

 only to consider what some of these characteristics are, 

 to be convinced that although they may have arisen in 

 early times, they cannot possibly have exis ed in the 

 times which were the earliest of all. Things may have 

 been done, and habits may have prevailed, when the 

 multiplication and dispersion of Mankind had proceeded 

 to a considerable extent, which cannot possibly have been 

 done, and which cannot possibly have prevailed when as 

 yet there was only a single pair of beings " worthy to be 

 called " man and woman, nor even when as yet all the 

 children of that pair knew themselves to be of one fam- 

 ily and blood. The word primeval ought, if it is to have 

 any definite meaning at all. to be confined to this earliest 

 time alone. It has already been pointed out, that on the 

 supposition thai trie condition of primeval man approxi- 

 mated to the condition of the lower animals, that con- 

 di ion could not have been nearer to, but must, on the 

 contrary, have been very much farther removed from the 

 condition of the modern savage. If, for example, there 

 ever was a time when there existed on one spot of earth, 

 or even on more spots than one, a single pair of human 

 beings, it is impossible that they should have murdered 

 their offspring, or that they shculd have killed and eaten 

 each other. Accordingly it is admitted that cannibalism 

 and infanticide, two of the commonest practices of sav- 

 age and of barbarous life, cannot have been primeval. 

 But this is a conclusion of immense significance. It hints 

 to us, if it does no more, that what is true of one savage 

 practice may possibly be true of otheis. 



( To be Continued ) 



ASTRONOMY. 



Comparison Stars :— Under this heading Mr. Dreyer, 

 in the last number of Urania, makes a most excellent 

 " Suggestion to Astronomers" upon a matter which, of 

 late, attracted some little attention. It is to be hoped 

 that other observers will follow the example set at the 

 Dunsink Observatory. Mr. Dreyer's "suggestion" is 

 as follows : 



" In spite of the numerous s'ar-catalogues in the hands 

 of observers of minor plants and comets, it frequently 

 happens that a well-determined place for a comparison 

 star cannot be found in any catalogue. Many s ars have 

 therefore to be re-observed, and much time is no doubt 

 lost by a number of observers, each having to determine 



the places of a few stars, which, if put together in one 

 working list could be observed by one person with but 

 little trouble. 



It would evidently be an advantage if an astronomer, 

 having at his disposal a good transit circle, would, for a 

 time, endeavor to determine the places of all the compari- 

 son stars recently used and requiring re-observation. 



In accordance with this scheme, I shall, until further 

 notice (with the concurrence of Dr. Ball) be glad to de- 

 termine with the Dunsirk Transit Circle the places of any 

 comparison s'ars north of — 20 Declination not found 

 in modern catalogues, and recently used in observations 

 of minor planets or comets. The mean places, based cn 

 the Fundamental Catalogue of the ' Astronomische 

 Gesellschaft,' will be worked out and published as scon 

 as practicable." 



THE SOLAR PARALLAX. 



M. Faye has recently communicated to the Paris Aca- 

 demy of Sciences {Comptcs Rendus Tome XCII., Ao. 8), 

 an interesting paper upon the actual state of our knowl- 

 edge of the sun's parallax. Remarking that there is no 

 other constant in science whose determination depends 

 upon such a large numb-r of entirely independent results, 

 he subdivides the various values assigned for the sun's 

 mean parallax, as follows : 



] 8.85" by Mars ( Tassini's methodi ..Newcomb 



Geometrical j 8.78 by Venus, 1769 I Halley's method) Powalky 



Methods j- 8.81 by Venus, 1874 " Tupman 



8.82 - 8.87 by Flora, (Galle's method) Galle 



J 8.79 by Juno " Lindsay 



Mechanical ^8 81 by the lunar inequality ( Liplace's methodi.. 



Methods >8.85 by the monthly equation of the earth Leverrier 



8.83" \ 8.83 by the perturberations of Mars and Venus. ..Leverrier 



Alethod' ! 3-799 ve ' oc ' ty °^ " gnt ' F ' zeau ' s method) Cornu 



* S 81" ) 8.813 " (Foucault's methodi Michelson 



In regard to the value 8.85 obtained by Cassini's method, 

 M. Faye says that Mars has always given values for the 

 solar parallax somewhat too large. The first value 8. "81 

 obtained by mechanical methods was calculated by 

 adopting for the coefficient of the inequality 125.2", the 

 mean between the result of Airy from the Greenwich ob- 

 servations, and that of Newcomb from the Washington 

 observations, taking for the moon's parallax 57' 2.7", and 

 for her mass -g-fers. By the second of the " mechanical 

 methods " Leverrier found 8. 95, which was afterwards 

 reduced to 8.85" by Stone upon correcting two slight 

 errors in the computation. The vaiue from the pertur- 

 bations of V'enus and Mars, assigned by Leverrier was 

 8.86", but one of the numbers requiring a small correc- 

 tion, it is reduced to 8.83". Michelson having overcome 

 all the difficulties in Foucault's method, found for the 

 velocity of light 2,999.40 kilom. ± 100 kilom. Using 

 Struve's constant of aberration the corresponding values 

 of the parallax are 8.799" an d 8.813", as above. The 

 general mean is 8.82", to which M. Faye attributes a prob- 

 able error of ± 0.016". Although each of the values 

 may be effected by systematic error, nevertheless, since 

 the causes of error are varied, and without the least pos- 

 sible connection, these errors must be to a great degree 

 eliminated, as well as the accidental errors. 

 The following conclusions are reached : 



1. That the physical methods are supetior to all 

 others, and should be adopted. 



2. That the value of the solar parallax, 8.813" ( DV ph)'~ 

 sical methods), is now determined to about T iy of a 

 second. 



3. That the seven astronomical methods converge 

 more and more towards that value, and tend to confirm 

 it, without equalling it in precision. 



This fact does not diminish, however, the great impor- 

 tance of observations upon the coming transit of Venus, 

 to which we can now bring to our aid the most effective 

 of photographic apparatus. W. C. VV. 



Washington, D C , Atr U 14. 1881. 



