March 15, 1907] 



SCIENCE 



433 



affected the velocity of its light, i\ would 

 affect the time required for the light to reach 

 us and would, therefore, change the apparent 

 position of the star at any instant. The 

 resulting distortion, which in some cases 

 would be great, has been carefully looked for 

 but has been found absent. 



Thus it is a fact founded on observation 

 that the velocity of light is independent of 

 the source. The meaning of this fact can be 

 made clear in the following way. Suppose an 

 observer imprisoned in a windowless box 

 which is thrown at random into space. Such 

 an observer, meeting no experimental diffi- 

 culties, could discover and accurately define 

 his speed through space by simply measur- 

 ing the velocity of light in various directions 

 within his enclosure, and, this, be it noted, 

 without any reference whatever to any outside 

 body. In general, he would find that light 

 travels faster across his box in one direction, 

 than it does in another, for, as we have seen, 

 the real velocity of light in space is not 

 affected by the motion of the light source 

 which he carries, and hence a change in his 

 motion would change the apparent velocity of 

 light within his box. 



Thus space possesses what we might call a 

 ' positional ' property, by means of which the 

 magnitude of any motion can be defined with- 

 out reference to any body in the universe, and 

 this motion is what physicists call ' motion 

 with respect to the ether.' 



There is an entirely different experimental 

 truth which leads to the same conclusion as 

 the above. 



It is a generally accepted truth that two 

 similarly charged bodies when moved side by 

 side have, superimposed upon their mutual 

 repulsion, an attraction which depends upon 

 the fact that when moving they act like two 

 parallel electric currents. This follows from 

 Eoland's classical experiment. 



Now if the two charges, stationary with re- 

 spect to each other, are considered alone in 

 space, it is evident that they furnish our 

 imprisoned experimenter another means of 

 •finding his motion relative to space, for the 

 strength of the above-mentioned attraction 

 depends only upon this absolute motion. This 



leads to the same conclusion as formerly, that 

 space has a ' positional ' property. 



Other evidence of a similar kind might be 

 given, but the above is sufficient to make it 

 clear that at least so far as it represents this 

 positional property of space the conception of 

 an ether is thoroughly legitimate. 



A being from a planet which possessed no 

 atmosphere, if he came to earth, might first 

 become conscious of our atmosphere through 

 feeling it set in motion relatively to himself 

 when he moved. In a somewhat similar way 

 the ether manifests itself, since we know it 

 through its motional property. 



We are conscious of matter only as a col- 

 lection of properties and one of these prop- 

 erties certainly is that it is capable of mark- 

 ing position. Therefore, the unknown reality, 

 which exhibits this positional attribute in 

 space as one of its properties, can .be said to 

 resemble matter to this limited extent at least, 

 and upon this sure foundation can fitly be 

 based the physicist's conception of an ether. 

 Daniel F. Comstock 



Cavendish Laboeatoey, 

 Cambridge 



NOTES ON ORGANIC CHEMISTRY 



GLYOXAL 



The properties of the substance known as 

 glyoxal, which is the simplest dialdehyde, 

 certainly do not accord with its formula, 

 OCHCHO. Some text -books try to evade this 

 difficulty by describing it as 'a hydrate,' but 

 this, to say the least, is hardly satisfactory. 

 The matter has been cleared up very com- 

 pletely by a recent paper of Harries and 

 Temme.^ Glyoxal is found to exist in four 

 forms : (1) Monomolecular glyoxal, OCHCHO, 

 which is unstable and is described more 

 fully below. (2) A trimolecular modification 

 (CjH20.)3, which is a colorless, pulverulent 

 solid, readily soluble in water. It forms no 

 characteristic derivatives, i. e., reagents give 

 with it only derivatives of (1). (3) Para- 

 glyoxal (CjHjOj)^,, a white powder, insoluble 

 in water. (4) The ordinary form of glyoxal, 

 now termed polyglyoxal CC,H„0.)„, which has 

 been known during the past fifty years. 



'Ber. d. Chem. Oes., 40, 165 (1907). 



