282 
THE VOYAGE OF H.M.S. CHALLENGEK. 
The largest forms hitherto discovered are from tropical or subtropical surface waters, 
e.g., Ethmodisci, Castracane= Coscinodiscus gazellae, Janisch, Coscinodiscus imperator, 
Janisch, Coscinodiscus praetor, Grove, Coscinodiscus nobilis, Grun., Coscinodiscus sol, 
Wallich. The most delicate species are especially tropical or subtropical, e.g., the peri- 
j)heral rotate rim of Coscinodiscus sol, many Rhizosolenise, and Chsetocerotidse ; this 
applies equally to the degree of tenuity of the siliceous test as a whole, to the nature of 
its ornamentation as determined by the difficulty of microscopical resolution, and to the 
siliceous appendages when present.^ This great development of Diatoms at the surface of 
Fig. 28.— Fnistale of Elhmodiscus wyviUeanus, Castracane Fia. 29. — Portion of Prustule of Ethmodiscus sp. (^). 
the sea takes place especially in brackish water, or in sea- water where the salinity is relatively 
low, as, for instance, in the Antarctic and Arctic Oceans, and in estuaries or off the mouths 
of great rivers. In the warm and salter waters of the ocean Diatoms are less abundant, 
and the frustules as a rule are much thinner than in the colder and less salt waters of 
the polar regions or in the warm brackish waters off continental shores in the tropics. 
In deep-sea deposits the remains of these pelagic Diatoms can generally be detected 
if a considerable quantity of the deposit be carefully examined. In some Pteropod and 
Globigerina Oozes and Coral Muds, however, no trace of Diatoms have been observed 
after the removal of the carbonate of lime from a large sample and a subsequent careful 
examination of the residue ; they are also extremely rare in, or absent from, some of the 
deep-.sca clays. In terrigenous muds, especially when near the mouths of great rivers, 
they frequently occur in great abundance. 
• Mr. John Rattray, M.A., F.R.S.E., a diatomist who ha.s examined many of the Challenger deposits, says in an 
ilS. letter to Mr. ^lurray; — “No dead tests are to be considered absolutely indestructible in time. Delicate Chieto- 
cerotidic are not found in oceanic deposits nor in geological strata. Coscinodisci disrupt along lines passing through, and 
not lietween, the hexagonal markings; the extremities of a Rhizosolenia sejiarate from the more destructible, spirally- 
ornamented, intermediate, cylindrical areas, an<l are alone preserved. In any one oceanic deposit, where the degree of 
■olrent power must lie the same for all forms, the degree of persistence varies directly as the strength of the siliceous 
parts, e.g., cingula arc less persistent than valves. On comparing the same species from widely-separated areas, the 
Isilance of evidence goes to show that more robust forms occur in more ]>olar areas, and more delicate ones in more 
ef|uatorial atvas. The dilferences oljservable are slight, but specimens of Coscinodiscus Icntiginosus, Coscinodiscus suhtilis, 
&c., point to their real existence.” 
