September 29, 1904] 



NA TURE 



531 



shape as the file in no way interfered with the experiment. 

 'J he substitution was of course unknown to the observer. 



I am obliged to confess that I left the laboratory with a 

 distinct feeling- of depression, not only having failed to see 

 a single experiment of a convincing nature, but with the 

 almost certain conviction that all the changes in the 

 luminosity or distinctness of sparks and phosphorescent 

 screens (which furnish the only evidence of «-rays) are 

 purelv imaginary. It seems strange that after a year's work 

 on the subject not a single experiment has been devised 

 which can in any way convince a critical observer that the 

 ravs exist at all. To' be sure the photographs are offered 

 as' an objective proof of the effect of the rays upon the 

 luminosity of the spark. The spark, however, varies greatly 

 in intensity from moment to moment, and the manner in 

 which the exposures are made appears to me to be especially 

 favourable to the introduction of errors in the total time of 

 exposure which each image receives. I am unwilling also 

 to believe that a change of intensity which the average eye 

 cannot detect when the ii-rays are flashed " on " and " off " 

 will be brought out as distinctly in photographs as is the 

 case on tjie plates exhibited. 



Experiments could be easily devised which would settle 

 the matter beyond all doubt ; for example, the following : — 

 Let two screens be prepared, one composed of two sheets 

 of thin aluminium with a few sheets of wet paper between, 

 the whole hermetically sealed with wax along the edges. 

 '] he other screen to be exactly similar, containing, however, 

 dry paper. 



Let a dozen or more photographs be taken with the two 

 screens, the person exposing the plates being ignorant of 

 which screen was used in each case. One of the screens 

 being opaque to the ii-rays, the other transparent, the re- 

 sulting photographs would tell the story. Two observers 

 would be required, one to change the screens and keep a 

 record of the one used in each case, the other to expose the 

 plates. 



The same screen should be used for two or three successive 

 exposures, in one or more cases, and it should be made 

 impossible for the person exposing the plates to know in 

 any way whether a change had been made or not. 



i feel verv sure that a day spent on some such experiment 

 as this would show that the variations in the density on the 

 photographic plate had no connection with the screen used. 



Whv cannot the experimenters who obtain results with 

 7i-ravs and those who do not try a series of experiments 

 together, as was done only last year by Cremieu and Pender, 

 when doubt had been expressed about the realitv of the 

 Rowland effect ? R. W. Wood. 



Brussels, September 22. 



Porpita in the Indian Seas. 

 During five voyages to and from the East, I have been 

 interested in watching for fand not always seeing) a species 

 of Porpita common in the Red Sea, on the coasts of India, 

 Cevlon, and the Malay Peninsula. From the deck of a 

 steamer the colony, only the flat disc of which is visible, 

 appears like a floating counter of bone or ivory. When 

 examined at close quarters it has a greyish metallic lustre, 

 and is seen to be surrounded with an aureole of azure 

 tentacles, the tips of which are green. So long ago as 

 i^7q ' Thomas Stevens appears to have remarked upon this 

 animal (though he did not recognise its animal nature) as 

 being one of the signs by which the vicinity of land might 

 be known on the Indian coasts. During the monsoon, even 

 in comparatively fine weather, this Porpita, so far as my 

 observations go, completely disappears from the surface. 

 It would seem to follow that the colony is an annual growth, 

 as it has no power of sinking, and very feeble, if any, means 

 of independent progression. This is borne out by an observ- 

 ation I was able to make on the shore at Colombo on 

 July 15 last. On that date, when the monsoon had already 

 been in progress for some weeks, the beach along the Galle 

 face, which is open to the full force of the monsoon, was 

 covered with biscuit-like discs, which I h.ad no difficulty in 

 recognising, from the sculpturing on their surface and the 

 characteristic appearance in cross-section, as those of 

 Porpita. They had quite lost their silvery appearance, and 



> See Beazley's " Voyages and Travels,' 



NO. 1822, VOL. 70] 



1903, p. 158. 



were very brittle ; no trace of the living tissues of the 

 animal remained. There were, however, large numbers of 

 other Siphonophora, too decomposed for even partial 

 identification (but obviously belonging to a different section 

 of the group), mingled with the discs. My friend Dr. J. H. 

 Ashworth tells me that he has observed much the same 

 thing in the Mediterranean with regard to Velella, and it 

 appears that Agassiz records having seen a broad blue band 

 of Velella along the shores of Florida, but I have not the 

 reference at hand. Nelson Annandale. 



Indian Museum, Calcutta, August 22. 



On van 't Hoff's Law of Osmotic Pressure. 



\'an 't Hoff imagines that a substance dissolved in a 

 fluid medium behaves as if it were in a vacuum, and so 

 exerts on the walls of the containing vessel a pressure which 

 is precisely that which it would exert were the solvent 

 imagined removed and the dissolved substance imagined 

 present in a gaseous form. 



The pressure thus exerted oy. the walls of the vessel is 

 called the "osmotic pressure." Many authors of great 

 mathematical repute have seriously questioned the correct- 

 ness of van 't Hoff's views, and they find it exceedingly 

 difficult to see how a dissolved substance can be present in 

 the solvent in a state similar to the gaseous state. 



For example. Prof. O. E. Meyer (" Kinetic Theory of 

 Gases," p. 367, Eng. trans., 1896) remarks ; — " . . . osmotic 

 pressure is not one of the phenomena w^hich the kinetic 

 theory of gases has to explain. I will also not conceal that 

 I do not think van 't Hoff's views of the kinetic nature of 

 osmotic pressure to be correct. For osmose does not arise 

 from the kinetic pressure of the dissolved substance, but 

 from quite different forces which cannot be neglected." 



I think, however, these authors have neglected an 

 important factor which would tend to make the dissolved 

 molecules behave as if in a vacuum, and so would tend to 

 give physical reality to van 't Hoff's views. 



The factor I allude to is the fact that different kinds of 

 molecules attract each other with enormously different 

 forces. For example, the molecules of carbon exert on each 

 other an enormous attractive force, as is shown by the 

 remarkable hardness and involatility of certain forms of 

 carbon. O.xygen, hydrogen, helium, and other molecules 

 have in comparison but a feeble molecular attraction. 



Consider a molecule -A in the midst of a swarm of other 

 molecules ; for example, a molecule in the interior of a 

 homogeneous liquid. Then if the molecule A be of the 

 same nature as the other molecules, each will exert the 

 same intensity of attractive force on the other, and so the 

 molecules will all be on an average symmetrically arranged 

 about A. The liquid will, in fact, have at every point a 

 symmetrical structure. If, however, the molecule A be 

 different in nature from the neighbouring molecules (as 

 occurs in the case of solution), two cases in general occur : — 



(i) The molecules of the liquid attract each other more 

 strongly than they attract the molecule A. 



(2) The molecules of the liquid attract each other less 

 strongly than they attract the molecule A. 



(i) In this case it is easy to see that under the influence 

 of the molecular forces the molecules of the liquid would 

 be drawn away from the molecule A (in precisely the same 

 way, and for a similar reason, that the molecules of quick- 

 silver are drawn away from glass), and so form about A 

 a sort of vacuum bubble ; and as A moves forward in the 

 liquid the molecules surrounding it would be drawn away, 

 and leave a free passage for A, which would thus behave 

 very much as if it were actually in a vacuum. Here, then, 

 van 't Hoff's conception becomes readily intelligible. 



(2) In this case molecules of the liquid would combine 

 with the molecule A to form an unstable compound, traces 

 of which are so often met with in solution ; and the com- 

 bination would proceed until the compound thus formed 

 exerted an attractive force on the neighbouring molecules 

 equal to or less than the force which the neighbouring 

 molecules exert on each other. 



When this occurs the case would resolve itself into case 

 (i) previouslv considered, the unit, however, being now not 

 the molecule A, but the molecular compound of which it 

 forms a part. 



