ANIMAL NATURE OF DIATOMEjE. 379 



the study of them, has the additional mischief of exalting, 

 in some degree, the value conventionally attached to 

 characters, and compels us to form a distinct family of 

 every genus. Such is the case with this attempt of 

 Kiitzing in respect to the last two genera with the name 

 of Meridiona, though they do not really differ from the 

 preceding genus Eunotia, or the following one Fragillaria, 

 except by a slight modification of form. Yet a similar 

 modification does not seem sufficient to distinguish 

 the two genera Meridian and Eumeridion. In general, 

 characters deduced from the form of single individuals, 

 certainly deserve a preference in comparison with those 

 furnished by associations of numerous ■ individuals into 

 polypariform bands ; but inasmuch as we do not know 

 the relation between outward figure and internal or- 

 ganisation, the subdivision of families founded only upon 

 this character seems premature, to say the least. 



6. Denticula. — Individua libera singidaria vel binatim 

 conjuncta, a latere primario oblonga I. linearia, secondaria 

 transversim striata I. costata ; strice pervies, validae. 



The morphological character of this genus is the pre- 

 dominance of the primary surfaces over the secondary, 

 and the continuity of striae crossing these latter. We shall 

 have an opportunity of returning to this particular subject 

 when treating of Surirella. At present it is sufficient 

 to observe, that whilst the first five species described by 

 Kiitzing, and figured in plate 17 of his work, show the 

 greatest affinity, not merely one to another, but also with 

 the subsequent genus, the two others, D. constricta and 

 D. undulata, display only how vague is the artifice of 

 this system which brings together objects so dissimilar, 

 and separates those which are truly allied. In these two 

 species there is not only wanting the predominance of 

 the primary over the secondary surfaces, but instead of 

 transverse striae, there are elevated costae, and in one of 

 them (T). undulata) even vittae are visible, as they are in 

 Surirella solea, to which it bears so great a resemblance, 



