28 



MEMOIRS NATIONAL ACADEMY OF SCIENCES. 



(Vol. XIV. 



The above analyses, it Avill be noted, cover, to a fair extent, the entire range in composition 

 of the stony meteorites. Although the number is small, it will nevertheless be not without 

 interest to average them, as was done in my paper of 1909," and by Farrington at a later date. * 

 In making this average I have adhered to the plan first adopted of considering only the actual 

 determinations of any particular constituent. Further, it has seemed advisable to rule out 

 a few cases in which the percentage amounts of any constituent were so high as to be considered 

 anomalous, as that of 2.39 per cent TiO, in the stone of Angra dos Reis, or 6.33 per cent Cr^Os 

 in that of Long Island. 



This elimination is of course open to serious objection, and the question may well be raised 

 as to the desirability of omitting entirely from the calculations the analyses in which such 

 anomalies occur. The only answer that can be given is that in so doing the total number 

 would be so reduced as to make any average of little interest. But it must be borne in mind 

 that the value of any average that can be made to-day lies not in its showing the actual average 

 composition, but rather in showing what has been done and inferentiaUy what remahis to be 

 done. 



The figures given m colmnn I are therefore averages of 53 analyses with the 15 exceptions 

 noted. In column II is shown for comparison the average composition of terrestrial igneous 

 rocks, and m III that of the lithosphere, as given by Clarke."^ In columns IV and V are repeated 

 the meteoric averages given in my previous paper and that of Dr. Fari'ington. 



Average composition of stony meteorites compared with terrestrial rocks. 



1 Average 



2 Avera?o 



3 Average 

 * Averapo 

 t Averatje 

 6 Avorage 

 ' Average 



of 46 determinations, 

 of 4'3 deierminations. 

 of 50 determinations, 

 of 41 determinations, 

 of 19 determinations, 

 of fi determinations, 

 of 33 determinations. 



8 Average of 49 determinations. 



9 Average of 44 determinations. 

 '(■Average of 15 determinations. 

 11 Average of 51 determinations. 

 '2 Average of 16 determinations. 

 '3 Average of 8 determinations. 

 '< Average of 5 determinations. 



a Amer. Journ. Sci., vol. 27, p. 469. 



c Bull. 491, U. S. Geol. Surv., 1911, p. 32. 



b Publ. 151 Field Mus. Nat. Hist., 

 Geol. Series, vol. 3, No. 9, 1911. 



