DIATOMACBiE — MBMBK ANACEO US FRONDS. 



299 



to the external ribbings. So far as is yet known, it is exclusively 

 marine. Nearly allied to it, is the beautiful genus Isthmia (Fig. 

 96), which is also marine, and which grows »ttached to larger 

 sea-weeds, the basal frustule being very conftnonly attached by a 

 stipes. In this, as in the preceding genus, the areolated structure 

 of the surface is very conspicuous (Fig. 78), both in the valves 

 and in the connecting " hoop ;" and this hoop, being silicified, 

 not only connects the two new frustules (as at h), until they have 

 separated from each other, but, after such separation, remains 

 for a time round one of the frustules, so as to give it a truncated 

 appearance (a, c). 



188. It is by the complete persistence of the siliceous hoop, 

 that the frustules are held together in that curious group of 

 Diatomacese, which consists of a few genera whose cylindrical 

 filaments bear a close external resemblance to those of Oonfer- 

 vaeece. The most important of these is Meloseira (Figs. 97, 98), 

 long since characterized as a plant by the Swedish algologist 

 Agardh, but taken from 



the Vegetable kingdom Fio. 97. Fio. 98. 



with other Diatoms by Prof. 

 Ehrenberg, who included 

 its species in his genus 

 G-allionella. Some of its 

 species are marine, others 

 fresh water ; one of the lat- 

 ter, the M. ochracea, seems 

 to grow best in boggy pools 

 containing a ferruginous 

 impregnation; and it is 

 stated by Prof Ehrenberg 

 to take up from the water, 

 and to incorporate with its 

 own substance, a consider- 

 able quantity of iron. The 

 filaments of Meloseira very 

 commonly fall apart at the 



slightest touch ; and in the naoseirasubjkxais. Mdosdravarians. 



infusorial earths, in which 



some species abound, the frustules are always found detached 

 ^Fig. 102, a a, d d). The meaning of the remarkable dift'erence 

 in the sizes and forms of the frustules of the same filaments 

 (Fig. 97, 98) has not yet been fully ascertained ; but it seems to 

 be related to the curious process of self-conjugation already 

 described (§ 178). The sides of the valves are often marked 

 with radiating strise (Fig. 102, d d); and in some species they 

 have toothed or serrated margins, by which the frustules lock 

 together. 



189. The Second tribe of Diatomacese, in which the frustules 

 are completely enveloped by a gelatinous or membranaceous 



