Mat is, 18S3.; 



SCIENCE. 



421 



a time will aid the passage of lymph ; but the anasto- 

 mosing vessels will carry the embolic ova as well as 

 the lymph. The corresponding glands will then, in 

 their turn, be invaded; and so on, until the entire 

 lymj)hatic system, connected directly or indirectly 

 with the vessel in which the parent worm is lodged, 

 becomes obstructed. 



"This, I believe, is the true pathology of the ele- 

 phantoid diseases: 1°. Parent Filaria in a distal lym- 

 phatic; 2°. Premature expulsion of ova; 3°. Embol- 

 ism of lymphatic glands by ova; 4°. Stasis of lymph; 

 5°. Regurgitation of lymph, and partial compensation 

 by anastomoses; 6°. Renewed or continued prema- 

 ture expulsion of ova; further embolisin of glands. 

 This process, according to the part of the lymphatic 

 system it occurs in, the frequency of its recurrence, 

 and its completeness, explains every variety of ele- 

 phantoid diseases." C. V. Riley. 



INTERNAL MOLECULAR ENERGY OF 

 ATOMIC VIBRATION.^ 



TuE object of this paper is to examine at length 

 the relative amount of energy which a molecule may 

 possess with respect to any small degree of freedom 

 of motion which its atoms may have as to each other. 

 The theorem of the virial is applied to this motion of 

 the atoms; and it is found, that in a molecule of a 

 perfect gas consisting of but two atoms, which are at 

 a mean distance, r, from each other, and which suffer 

 a small displacement whose mean maximum ampli- 

 tude is 6r under the action of elastic forces, the energy 

 of atomic vibration will be to that of translation par- 

 allel to any assumed direction in space as 6r to ?•. It is 

 further shown that this result is of such a character 

 as not to be restricted to molecules of two atoms 

 merely, nor to atoms which are attracted toward their 

 mean position by forces varying simply as the first 

 power of the displacement; so that the result arrived 

 at is of a general nature which may be stated thus : 

 the energy of interatomic vibration depends upon the 

 atomic displacement within the molecule, and in 

 such a way, that, when this displacement is a vanish- 

 ing quantity compared with the dimensions of the 

 molecule, then this energy of internal vibration is 

 a vanishing quantity compared with the energy of 

 motion of the molecule as a whole. 



This result is in confirmation of the results ob- 

 tained by the author in his previous paper upon ' An 

 extension of the theorem of the virial,' '-^ etc., in 

 which he expressed the opinion that the results there 

 obtained led to the conclusion, that " in case partial 

 constraints not amounting to the loss of entire de- 

 grees of freedom are introduced, the energy will no 

 longer be equally distributed among the co-ordinates, 

 but will be influenced by their constraints." 



This being in direct contradiction to the conclusions 

 which have been deduced by Boltzmann and by Wat- 

 son from the discussion of the distribution of energy 

 by the method of generalized co-ordinates, an exami- 

 nation, is made of the point in this hitherto accepted 

 theory from which the contradiction arises, and an 

 error is pointed out in the method of employing the 

 fundamental expression for the distribution of veloci- 

 ties. The error is of this nature: the law expressing 

 the most probable distribution of velocities with 

 respect to any single co-ordinate is the same as that 

 of the most probable distribution of errors of obser- 



1 Abstract of a paper upon a further extension of the theorem 

 of the virial to the internal molecular energy of atomic vibration. 

 By H. T. Eddy, Ph.D., Cincinnati. Read before the Section in 

 physics and chemistry of the Ohio mech. inst. April 26, 1S83. 



- Sc. proc. Ohio mech. inst.t March, 1883; Science, p. 65. 



vation, and contains a single arbitrary constant, to be 

 determined by the observations themselves. It has 

 been assumed that this constant is the same for each 

 co-ordinate, which is, in effect, assuming the very 

 point to be proved. It is here pointed out, that doing 

 this commits an error of the same nature as is done 

 in assigning equal weights to unlike observations 

 without first showing that their weights are equal. 



The computations made by means of the virial 

 show conclusively that the mean energy (i.e., the 

 weight) is not at all the same for one degree of free- 

 dom as for another; and, in order to find how one is 

 related to another, it will be necessary to take ac- 

 count of the forces acting, as has been done in this 

 paper and in the previous one. 



This extension of the theory leads to numerical re- 

 sults in close accordance with observed values of the 

 specific heats of gases, and their ratio, without pre- 

 vious knowledge of these quantities for any gas; thus 

 computing these quantities for the first time solely 

 from the general equations of mechanics. 



ON THE DEVELOPMENT OF CHLORO- 

 PHYLL AND COLOR GRANULES. 



The view has been generally entertained, based 

 largely on the admirable investigations of Arthur 

 Gris, that chlorophyll-granules are produced by direct 

 differentiation of the protoplasm of assimilating cells. 

 Led by his study of certain protoplasmic bodies in 

 the cells where nutritive matters are stored for future 

 use, and following out a suggestion made by Schmitz 

 in his recent work relative to the assimilating bodies 

 in certain Algae, A. F. W. Schimper (Bolaii. zeit., 

 Feb. and March, 1SS3) has made a detailed examina- 

 tion of the origin of chlorophyll-granules, which in- 

 dicates that the views of Gris are erroneous. At the 

 points of growth examined by him, Schimper uni- 

 formly found that well-formed granules already exist, 

 and that, from subsequent division of these, all the 

 cblorophyll-gi'anules are produced. From these, and 

 not, as heretofore believed, from the differentiation 

 of the protoplasmic mass in the cell, arise the gran- 

 ules which later, under the influence of light, take 

 on their characteristic color. One of the most in- 

 teresting cases reported by him is that of Azolla. 

 The point of growth at the root contains bright 

 green chlorophyll-granules about as large as those in 

 the older parts, antl in these granules the process of 

 division is to be distinctly traced. 



In those points of growth where the tissues are as 

 yet free from color, he has been also able to follow 

 the division, step by step, up to the prodtiction of 

 complete green granules. The bodies from which 

 the granules are produced are present, likewise, in 

 all points of growth of seedlings. Just here is 

 found the most interesting feature of this investiga- 

 tion. From these bodies, which he well terms ' plas- 

 tides,' come three classes of protoplasmic bodies, 

 somewhat resembling one another in shape: namely, 

 1, the chlorophyll-granules, or chloroplaslldex : 2, 

 the starch-formers, which, with the allied white or 

 colorless bodies, he calls leucoplastides ; and 3, the 

 bodies which possess colors other than green (for 

 instance, the granules in petals and the like), to which 

 he gives the name chromoplafstides. To illustrate 

 this from a single case, we will allude to Impatiens 

 parvifiora. The very transparent cells at the point 

 of growth contain plainly visible leucoplastides. In 

 cells of the same age they are of the same size, often 

 constricted, always sharply defined. These can be 

 traced by plain transitions into chloroplastides on 

 the young stem and the zone of forming leaves, 



