14 



SCIENCE 



[N. S. Vol. LII. No. 1331 



number of spines. Later the scales grow still 

 smaller, becoming like shagreen; bony plates 

 appear, while the spinous dorsal, ventral fins 

 and the gill openings undergo reductions. 

 Later the spinous dorsal and the ventrals dis- 

 appear altogether, the teeth coalesce into two 

 in each jaw and finally into one in each jaw; 

 this series finding its extreme in the head-fish 

 (Mola), in which the body is deeper than 

 long and seems to be simply a great head with 

 a fin behind it. 



Turning in another direction, the spinous 

 fins disappearing, the body is covered with 

 bony plates, and these finally interlock with 

 each other, foi-ming a complete bony box 

 absolutely immovable. Species thus provided 

 are known as trunk-fishes {Osiracion), and in 

 these the bony plates sometimes extend them- 

 selves into spines, especially on the head, 

 which thus acquires a fantastic appearance. 



Similar changes are found in other groups, 

 the general rule being extreme specialization 

 of a particular organ, producing its expansion 

 and high development, ultimately followed by 

 its reduction and final disappearance. In each 

 group the species most normally formed are 

 the earliest to appear in geological history, 

 while primitive forms often linger with the 

 others to the present time, most of these 

 groups having their origin in the Eocene. 

 Thus many of these early types, even the earl- 

 iest, remain to the present day, showing ap- 

 parently that non-specialization, ultra-special- 

 ization and loss of structure are all of second- 

 ary importance in the struggle for existence, 

 and that they are conditioned on something 

 else, a law not yet understood. 



David Starr Jordan 



Stanford TJNivEBSirT 



EINSTEIN'S THEORY AND SHIFT OF 

 SPECTRUM LINES 



According to Einstein's theory, as I under- 

 stand it, any time piece, as e. g., a vibrating 

 atom, automatically goes slow if placed in a 

 strong field of gravitation; also the effect of 

 a gravitational field is not to be distinguished 

 from inertia effects in any accelerated motion. 

 A particular illustration is that of a clock 



moving in a circle. It would seem that there 

 should not be much difference between the 

 effects of an acceleration produced thus and 

 that produced by magnetic or electric fields. 



In the Stark effect, where the radiation 

 from atoms in a strong electrical field is 

 studied, it is probable that some of the 

 radiating atoms are in a charged state. If 

 one computes the acceleration of a hydrogen 

 molecule with one unit of charge in a field of 

 20,000 volts per cm., without considering the 

 dragging effect of other molectdar fields, one 

 finds it to be of the order of 10^* cm. per sec. 

 per sec, much greater than the value of g at 

 the sun, so that if the atom could radiate in 

 this state very large displacements in the 

 spectrum lines should be expected, amounting 

 to the appearence of new lines. Even if the 

 atom as a whole is neutral, yet because of the 

 nonliomogeneity of the field and the distance 

 between the positive and negative constituents 

 of the atom, considerable accelerations are to 

 be expected which will be larger according as 

 an electron is farther removed from the 

 central nucleus. It is interesting then to re- 

 call that Stark found in several instances a 

 displacement of his central image towards the 

 red end of the spectrum, and found the com- 

 ponents unsymmetrically placed, also that in 

 a number of cases entirely new lines have 

 been found. 



If the preceding point of view is correct, 

 then in any ease of luminosity in a gas, since 

 during " collisions " the atoms are evidently 

 in strong fields of force, a slight displacement 

 of the center of gravity of a line towards the 

 red would appear, and this would increase 

 with the pressure as in the common pressure 

 effect. The explanation of the pressure shift 

 as due to the action of adjacent molecular 

 field? has been given, and according to Ein- 

 stein's theory that would bring it into accord 

 with the general relativity views of time. 

 The difficulty of distinguishing between the 

 pressure effect and the one predicted by Ein- 

 stein in the sun has been noted. Is it not 

 possible that this lies in the nature of things, 

 the difference being that while all matter is 

 subject in the same degree to gravitation, the 



