iU 



KNOWLEDGE. 



September, 1911. 



KlGURE IQ. p. haUicuiii showiiiL; the fibrils 

 and thr double structure of the vah e. X 1750. 



All the work fit^urcd here 

 hitherto was done upon 

 diatoms mounted drw Later 

 the writer was able to com- 

 plete his evidence, of a unity 

 of structure, upon other 

 d i a t o m s m o u n t e d i n a 

 medium. I"i,L;ures 20 and 

 2] exhihit the same struc- 

 ture of tilirils upon a small 

 CosciiKuiiscus so mounted, 

 and luMui; of sufficient 

 thickness to allow the separ- 

 ate layers to be distinguisluil 

 by an oil immersion. Li 

 one [jrint the picture is much 

 more crowded than apjieared 

 under the microscope, where 

 the fine adjustment could be 

 used. Onl\- those who prac- 

 tice photo-micrography can 

 appreciate the difference be- 

 tween seeing things in the microscope, and putting 

 the same features \i\i<iK- upon a plate. X'isually 

 the fibrils were unravelled, la\er after la\er, as one 

 might unravel some woven matirial. b'igure 21, 

 taken at a slightly different level, shows two fibrils 

 projecting into space, well separated from evervthing 

 at the sides, above and below. E\'i(lentl\- tlie fibrils 

 seen in the two prints, and not Mr. .Slack's beads, 

 form the sides of the he.xagoiis of the middle la\'er 

 of this diatom. 



It is not assumed here that all diatom structure 

 must be made up of similar fibrils. It is alwa\s 

 dangerous to generalise from too little data. On the 

 other hand the evidence is positive that much of the 

 structure is, and there seems nothing to preclude the 

 remainder from being formed of the same kinds of 

 units. How these fibrils are deposited upon the 

 true membrane of the lining vegetable cell is 

 another matter. 



i^t^'iti- 



I'lGCRE 20. .A small Co^^cuiotiisciis \al\e 



mounted in balsam, showing the same kind 



of fibrils forming the hexagons, X 1750. 



This article would not be complete without 

 reference to Dr. Henry \'an Heurck's more recent 

 work upon diatom structure, with the new a[)Ochro- 

 matic objective of 1-63 N..\. Plate 11 of the last 

 edition of Carpenter was prepared by Dr. \'an 

 Heurck especialh' for that work. The description 

 reads: — "Thisiilate has a two-fold purpose. It is 

 designed, first, to justif\' the opinions held b\' Dr. 

 \'an Heurck u[)on the structure of the vah'es of 

 diatoms, and also to show how the usual micro- 

 scopical tests [iresent themselves when examined w ith 

 the new objecti\e . . . lately constructed by the firm 

 of Zeiss." 



The editor (Dr. Dallinger) thus describes Dr. 

 \'an Heurck's views. " He concludes that diatom 

 valves consist of two membranes or thin films and an 

 intermediate laver, the latter being pierced with 

 openings. The outer membrane is delicate and may 

 be easilv destro\'ed b\' acids, friction, and the 

 several processes of cleaning. 

 When the openings or aper- 

 tures of this interior portion 

 are arranged in alternate 

 rows, thev assume the hex- 

 agonal form: when in straight 

 rows then the openings are 

 square or oblong." 



Dr. Dallinger continues: 



It is, howe\'er, due to Mr. 



T. !•". Smith, who worked at 



this subject for years, to say 



that he long maintained this 



\iew In Plate 1, 



I'igure 1, we ha\e a photo- 

 graph of his, showing the 

 inside of a valve of P. 

 iiui^iildtuin magnified one 

 thousand se\en hundred and 

 fifty diameters, and exhibit- 

 ing the " postage stamp " 

 while in I'iijure 2, 



fracture 



■'^m^^i^^ 



■IGIIKK 



21. The same taken at a little 

 different level, X 1750. 



