4 P. T. CLEVE, SYNOPSIS OF THE NAVICULOID DIATOMS. 



In carrying out tliis work I have been kindly assisted by several diatomists, who sent me 

 for examination slides from their collections. Among them I name with recognition Prof. .1. Brux, 

 jMr KiNKER, Mr. Le-Tourneur, Mr. .1. D. Möller, Mr. P. Petit, Dr. Rae, Mr. Thum, Mr. E. Weissflog, 

 Prof. Van Heurck and Mr. Ward. I am indebted to Mr. Julien Deby, who sent from his gigantic 

 collection everything of interest to me, and to Mr. E. Grove, who not only sent me a large nuni- 

 ber of slides, bnt also took the trouble of revising the manuscript and tbe proofs. Materials from 

 difFerent parts of the world have been sent by Dr. Aurivillius (from Java), Mr. Beddome (from 

 Tasmania), Mr. DusÉx (from Cameroon). Captain G. C. Eckman (marine mud from the Atlantic and 

 Mediterranean), Prof. Lagerheim (from Ecuador) and Dr. Norstedt (from Australia. New Zealand 

 etc). To Mr. Comber 1 am indebted for many tine photographs of several forms of interesting struc- 

 ture. To all these gentlemen I give my best thanks. 



On the value of the characteristics. 



It may be stated as an axiom that those characteristics are of the greatest importance which 

 occiir in the greatest number of forms, and on the other hand that the characteristics which occur 

 in some, but not in all, of a number of forms otherwise nearlj' related are of less importance. The 

 older authors as a rule attached but little importance to the more constant characteristics, but on 

 the contrary gave much attention to trifling differences, because the latter are frequentlj' more 

 conspicuous than the former. They founded not only species but genera and divisions on charac- 

 teristics, which were actually subject to v^ariation in species otherwise identical. I will in the 

 following pages treat of all characteristics which have been used for the description of species and 

 genera. 



Habif of Ii fe and groivth. Most diatoms live in a free state, Hoating in the water, but many 

 are attached to solid bodies in the water, some are stipitate on gelatinous stalks, and others enclosed 

 in gelatinous tubes or masses of difPerent shape and consistence. A careful examination shews that 

 many of the attached or enclosed forms also occur in a free state; and that there are frequently 

 very slight differences between species, which live attached or enclosed, and others which never 

 occur in such a state. On the other hand forms, which are stipitate, or enclosed in tubes, belong 

 to the most different tv'pes. Eor these reasons I regard as a characteristic of very little im- 

 portance the mode of occurrence in free or attached state. At least, genera and species should 

 not be founded on such characteristics alone. Genera, such as Schizonema, Endostauron, Endosic/ina, 

 Encyonema, Cocconenia, are in my opinion not acceptable. > 



Size. The limits of the dimensions of the forms of each species are in most cases pretty 

 definite, the larger forms of each species being as a rule twice as large as the smallest. Still, in 

 some cases the variability is more extensive and the followings may be cited as species in which 

 the dimensions of the forms are siibject to very considerable variation: Amphora ovalis. Achnanthes 

 brevipes, Pinnularia viridis. 



Form of the frustule. Most frustules of the Raphidiefe are straight, and nearly symmetrical. 

 with the longitudinal and transverse axes, but a great number of forms are in some or other 

 respects asymmetrical. A flexure of the frustule along the longitudinal, or the transverse axis. 

 occur in the old genera Cocconeis and Achnanthes which have been distinguished hitherto principally 

 by this characteristic. As to Cocconeis it seems probable that the flexure may be derived from the 



' The gelatinons substance of the stipes of Ächiwnfes longipcs is intensely stained by ha-matoxyline, and no con- 

 tinuation of the snbstance of the stipes is visible in the interiör of the frustule. On the other hand the stipes is not stained 

 by Congo-red, methyle-green, eosine, and methyléne-blue. The gelatinous tubes of Navioula (Schizonema'1 mollis are stained 

 by hsmatoxyline, fuchsine, methyle-green and saffranine. but not by eosine. The tubes of Amphiplenra (.Berkelya) Dilhvynii 

 are stained by hsematoxyline and methyle-green, but not by Congo-red. The gelatinous envelopes of Mastologia are stained 

 by methyle-green, fuchsine and saffranine. but not by eosine and Congo-red. 



