142 



THE AMERICAN MONTHLY 



[August, 



with good objectives, can be distinctly 

 seen in the midst of the pores ; this 

 structure is common in diatoms. I 

 have a large specimen of this species 

 in which a very fine and close punc- 

 tation seems to exist on the internal 

 face of the valves. The section of 

 Cose, excentriciis (or more likely Cose, 

 sytnbolophord), given by M. Prinz, ap- 

 pears to be very accurate ; the canals 

 of the pores are in great part shown to 

 not reach the internal face of the 

 valves. 



M. Prinz,* after reading the criti- 

 cisms of Mr. Grunow, called to mind 

 that his observations were made upon 

 material provided by Mr. Mauler, at 

 the time of the publication of a me- 

 moir of Mr. J. Deby " On the Micro- 

 scopical Appearances of Diatoms, " and 

 of a note by Dr. Kaiser on the pho- 

 tographs of Pleurosigma angulatum. 

 He said : '' I hoped, by means of thin 

 sections of rocks placed at my dispo- 

 sition, to verify the observations of 

 those authors. I thought it would suf- 

 fice to announce the results obtained 

 by this mode of preparation to induce 

 diatomists to take part in the study of 

 certain delicate questions of organo- 

 graphie and of structure. The few 

 details into which I then entered (and 

 of which I hope to be pardoned for 

 the imperfections due to imperfect no- 

 tions upon the subject) had no other 

 aim. 



" I may be permitted to remark that 

 my honorable opponents have drawn 

 their principal arguments from facts 

 observed under other conditions than 

 mine, and resulting from the exami- 

 nation of different species than those 

 I have described ; which leads to sup- 

 pose that I have applied the results of 

 my researches to the structure of all 

 diatoms. I believe, on the contrary, 

 that many species present details of 

 structure which do not correspond to 

 the conformation of other species re- 

 sembling them in appearance. 



" The details of structure of the Pleu- 

 rosigma a?igie/aiu?n, appear to me too 



* Bull, de la Soc. Beige de Microscofie, 



complicated and of a nature too delicate 

 to decide the question. The thinnest 

 sections of this species have at least 

 two series of pearls or pores. The sili- 

 cious frustules which contain these de- 

 signs show in section the image of two 

 continuous traces, one interior and the 

 other exterior. It is this that led Mr. 

 Flogel to admit the existence of cham- 

 bers closed by these membranes. 

 This image is produced with all sec- 

 tions, even with those of large species, 

 but that aspect disappears when the 

 section is sufficiently thin and when 

 it is examined .with high-powers. It 

 is then seen that the black lines, rep- 

 resenting the membranes, are discon- 

 tinuous and interrupted by bright spots 

 representing openings. It is under- 

 stood that to study these details, 

 the thinness of the section should de- 

 pend upon the fineness of the mark- 

 ings to be resolved. The greater 

 part of the fine work of M. Flogel has 

 been again put in question by the ob- 

 servations of Mr. Miiller, who has 

 shown that the sections obtained by 

 that savant were too thick and that the 

 closed chambers which -he. described 

 should have been open, since they be- 

 came filled by immersion in balsam. 

 Miiller, however, only admits their 

 opening on tlie e-xterior side. 



" The impressions in collodion do not 

 seem to havp given the authors who 

 have employed that process the results 

 expected. On the other hand, the 

 submersion of valves has likewise fur- 

 nished facts which fail of concord- 

 ance. 



" The figures obtained by photog- 

 raphy are, as the theory of Prof. Abbe 

 shows, the reproduction of illusory 

 images, due to diffraction fringes, at 

 least for diatoms of which the markings 

 are so arranged as to give place to 

 phenomena of this kind. The exam- 

 ination of valves by reflected light, as 

 proposed by Mr. Deby, proves nothing 

 more than the examination by trans- 

 mitted light, for it is not possible to 

 eliminate by this process, the causes 

 of error due to phenomena of diffrac- 

 tion. 



