January 24, 1896.] 



SCIENCE. 



113 



The first method employed consisted in 

 smearing frog's lymph on a slide, plunging 

 it into mercuric solution and passing it 

 through different stains. Such films gave 

 a granulo-reticular appearance strongly 

 stained and quite similar to many ordinary 

 cell protoplasms. The absence of any con- 

 trol as to thickness of film, however, makes 

 this method inapplicable to rigid compari- 

 son with appearances of protoplasm in sec- 

 tions of known thickness. In order to 

 meet this difiiculty, although possibly intro- 

 ducing others, the author inserted small 

 bits of dry pith into the lymph sac of a 

 frog, and after these had become saturated 

 with lymph they were removed and with 

 similar sized bits of other tissues were 

 passed through various histological pro- 

 cesses and sectioned in paraffin. Thus sec- 

 tions of tissue and of lymph coagulum, fil- 

 tered through the walls of pith cells, were 

 obtained, of equal thickness and compara- 

 ble in every way. 



Compared thus with cells of nerve, mus- 

 cle and gland the chief result is that lymph 

 furnishes to empty pith cells a histological 

 content strikingly similar to certain struc- 

 tures usually ascribed to protoplasm. Ee- 

 cently Fischer, by injecting pith with chem- 

 ically prepared solutions of proteids, pep- 

 tones, et at., proved that a number of re- 

 agents precipitated these proteids in the 

 form of granules not to be distinguished 

 from Altman's ' Elementarorganismen.' It 

 thus becomes manifest that the granular 

 factor in cell protoplasm may be readily ac- 

 counted for as a simple artefact formed 

 from solutions and not necessarily as pre- 

 formed in the cell. In the author's experi- 

 ments on lymph in which osmic acid, Flem- 

 ming's solution, mercuric chloride, gold 

 chloride and alcohol were used for harden- 

 ing, the character of the precipitate was 

 chiefly reticular or alveolar. In alcohol 

 and mercuric chloride this is fine and ap- 

 pears under ordinary powers as vacuolated 



granular protoplasm. In osmic solutions it 

 is coarsely alveolar with dense accretions of 

 stained matter at the angles of the alveoli. 

 Gold chloride gives a striking fibrillar retic- 

 ulum with frequent sharply defined gran- 

 ules, resembling the varicosities and end 

 balls often described in connection with 

 nerve fibrils. 



A number of stains have been tried. The 

 carmines and hsematoxylins are strongly 

 retained, as are most of the anilins, eosin, 

 fuchsin and nigrosin, and even methyl blue 

 and safranin are retained quite strongly. 

 Comparison with cells of different tissues 

 prepared side by side with the lymph from 

 the same frog would thus indicate that a 

 considerable proportion of the substance 

 stained in the cell protoplasm can not be 

 differentiated from lymph by the stains 

 thus far employed. It is true that identity 

 of staining cannot be taken to prove iden- 

 tity of substance ; but until other methods 

 of analysis prove either identity or difier- 

 ence, we must admit the possibility that a 

 large factor in what is ordinarily described 

 as the granulation or reticulation of cell pro- 

 toplasm may be simply precipitate in the 

 cell of lymph common to the whole body. 

 Until such analysis is made, further work 

 upon the finer ' structure ' or even on the 

 ' content ' of the so-called ' protoplasm ' can 

 have little permanent value. A point of 

 special importance is that the nucleus stains 

 by almost all methods in a way to differ- 

 entiate it sharpljr from lymph precipitate. 

 These reactions would disprove all ideas 

 tending to make the nucleus a lymph space 

 in the cell. » 



9. C. F. Hodge (for J. E. Slonaker) : 



Demonstration of the comparative anatomy of 



the area and fovea centralis. 



Methods for preserving the eye and for 



the demonstration of the retina in the eye 



as a whole and in microscopical sections 



were briefly discussed and a large number 



