1866.] CARPENTER EOZOON CANADENSE. 225 



almost sure to present itself. This constant relation between the 

 development of the " intermediate skeleton" and that of the " canal- 

 system " is in snch precise accordance with what I have uniformly 

 found to exist in Poraminifera*, as to add another fact of no mean 

 significance to our growing accumulation of evidence. 



Having dwelt sufficiently in my former note on the characters which 

 I regard as essentially Foraminiferal, I need not here do more than 

 recapitulate them as follows : — 1. The constant segmental division 

 of the siliceous layers, indicating a series of chambers only partially 

 divided from each other, as in Carpenteria, but occasionally sepa- 

 rated completely by interposed calcareous septa. 2. The perforation 

 of these septa by several passages of communication between the 

 chambers, as in Cijclodiipeusf. 3. The substitution of the acervuline 

 piling-together of the chambers, in the latter stages, for the regularly 

 storied arrangement of the earlier ; a change very common among 

 Foraminifera. It is further not a little significant that the siliceous 

 segments in this acervuline portion have the same average dimensions 

 as those of which the subjacent laminse are composed. The uni- 

 formity with which the acervuline mode of aggregation succeeds the 

 lamellar could scarcely be expected if these masses were of purely 

 mineral origin. 4. The distinctness of the proper wall of the cham- 

 bers from the intermediate skeleton ; a feature especially charac- 

 teristic of the higher Poraminifera. 5. The precise conformity in 

 the structure of the former to that of the Nummuline Poramini- 

 feraj. 6. The constant proportion between the development of the 

 latter and that of the canal-system. 7. The correspondence in size 

 between the elementary parts of Eozoon and those of the Porami- 

 nifera to which it is most nearly related. This is a feature of no little 

 importance, when we take into account the extreme variability in the 

 size of the elementary parts of Mineral aggregations §. It is almost 



* Introduction to the study of the Foraminifera, p. 63. 



t The explanation of these passages offered by Messrs. King and Rowney, is 

 based on the representation of them given in the somewhat diagrammatic figure 

 illustrating my former note (Plate VIII. fig. 2). That it is totally inapplicable 

 to the case, becomes at once apparent on miscroscopic examination of the large 

 number of examples I now possess of this very characteristic feature. 



\ It is urged by Profs. King and Rowney as an objection to the Foraminiferal 

 doctrine, that this supposed tubular wall is found on the under or attached side 

 of each lamella, as well as on the upper or free surface, and that it is incon- 

 ceivable that a layer thvis perforated for the passage of pseudopodia should be 

 formed where there is no exit for them. The objection only shows their want 

 of acquaintance with the fact that many Foraminifera (both recent and fossil), 

 having perforated shells, habitually grow affixed to seaweeds, corals, shells, &c. ; 

 and that the attached side possesses the characteristic tubular structure no less 

 than the free. I beheve that in all these cases there intervenes in the living 

 state a thin layer of sarcode between the shell and the subjacent surface, the 

 evidence of whose presence -in Eozoon is afforded by the continuous siliceous 

 lamellae already referred to as often connecting the outer ends of the aciculi of 

 the asbestiform layer. 



§ The only other objection of any weight advanced by Messrs. King and 

 Rowney, is the fact that a chambered structure and asbestiform layers are some- 

 times met with in the midst of the thick purely serpentinous layer which usually 

 underlies the masses of Eozoon. Now if such be really the fact, it would only 

 prove the persistence of bands of the original structure in parts which had been 



