A HALF-CENTURY OF SCIENCE. 197 



of carbon in some form. A photographic spectrum of the comet re- 

 cently visible, obtained by the same observer, is considered by him 

 to prove that nitrogen, probably in combination with carbon, is also 

 present. 



No element has yet been found in any meteorite, which was not 

 previously known as existing in the earth, but the phenomena which 

 they exhibit indicate that they must have been formed under condi- 

 tions very different from those which prevail on the earth's surface. 

 I may mention, for instance, the peculiar form of crystallized silica, 

 called by Maskelyne, asmanite ; and the whole class of meteorites, 

 consisting of iron generally alloyed with nickel, which Daubree terms 

 holosiderites. The interesting discovery, however, by Nordenskjold, 

 in 1870, at Ovifak, of a number of blocks of iron alloyed with nickel 

 and cobalt, in connection with basalts containing disseminated iron, 

 has, in the words of Judd, " afforded a very important link, placing 

 the terrestrial and extra-terrestrial rocks in closer relations with one 

 another." 



We have as yet no sufficient evidence to justify a conclusion as to 

 whether any substances exist in the -heavenly bodies which do not 

 occur in our earth, though there are many lines which can not yet be 

 satisfactorily referred to any terrestrial element. On the other hand, 

 some substances which occur on our earth have not yet been detected 

 in the sun's atmosphere. Such discoveries as these seemed, not long 

 ago, entirely beyond our hopes. M. Comte, indeed, in his " Cours de 

 Philosophic Positive," as recently as 1842, laid it down as an axiom 

 regarding the heavenly bodies, that " nous concevons la possibility de 

 determiner leurs formes, leurs distances, leurs grandeurs et leurs 

 mouvements, tandis que nous ne saurions jamais etudier par aucun 

 moyen leur composition chimique ou leur structure mineralogique." 

 Yet within a few years this supposed impossibility has been actually 

 accomplished, showing how unsafe it is to limit the possibilities of 

 science. 



It is hardly necessary to point out that, while the spectrum has 

 taught us so much, we have still even more to learn. "Why should 

 some substances give few, and others many, lines ? Why should the 

 same substance give different lines at different temperatures ? What 

 are the relations between the lines and the physical or chemical prop- 

 erties ? We may certainly look for much new knowledge of the hid- 

 den actions of atoms and molecules from future researches with the 

 spectroscope. It may even, perhaps, teach us to modify our views of 

 the so-called simple substances. Prout long ago, struck by the re- 

 markable fact that nearly all atomic weights are simple multiples of 

 the atomic weight of hydrogen, suggested that hydrogen must be the 

 primordial substance. Brodie's researches also naturally fell in with 

 the supposition that the so-called simple substances are in reality com- 

 plex, and that their constituents occur separately in the hottest regions 



