FORAMINIFERA AND RADIOLARIA. 105 



been formed by the coalescence of the pseudopodial filaments that passed 

 through the tubulated lamellae. 



495. In the fossilized condition in which Eozoon is most commonly 

 found, not only the cavities of the chambers, but the canal-systems to 

 their smallest ramifications, are filled up by the siliceous infiltration 

 which has taken the place of the original sarcode-body, as in the cases 

 already cited ( 487 note)\ and thus when a piece of this fossil is sub- 

 jected to the action of dilute acid, by which its calcareous portion is 

 dissolved-away, we obtain an internal cast of its chambers and canal- 

 system (Plate xvii., fig. 2), which, though altogether dissimilar in ar- 

 rangement, is essentially analogous in character to the ' internal casts' 

 represented in Figs. 328, 332. This cast presents us, therefore, with a 

 model in hard Serpentine of the soft sarcode-body which originally occu- 

 pied the chambers, and extended itself into the ramifying canals, of the 

 calcareous shell; and, like that of Polystomella ( 487), it affords an 

 even more satisfactory elucidation of the relations of these parts, than 

 we could have gained from the study of the living organism. We see 

 that each of the layers of serpentine, forming the lower part of such a 

 specimen, is made up of a number of coherent segments, which have 

 only undergone a partial separation; these appear to have extended 

 themselves horizontally without any definite limit; but have here and 

 there developed new segments in a vertical direction, so as to give origin 

 to new layers. In the spaces between these successive layers, which 

 were originally occupied by the calcareous shell, we see the ' internal 

 casts' of the branching canal-system; which give us the exact models of 

 the extensions of the sarcode-body that originally passed into them. 

 But this is not all. In specimens in which the nummuline layer consti- 

 tuting the 'proper wall' of the chambers was originally well preserved, 

 and in which the decalcifying process has been carefully managed (so as 

 not, by too rapid an evolution of carbonic acid gas, to disturb the ar- 

 rangement of the serpentinous residuum), that layer is represented by a 

 thin white film covering the exposed surfaces of the segments; the super- 

 ficial aspect of which, as well as its sectional view, are shown in fig. 2. 

 And when this layer is examined with a sufficient magnifying power, it 

 is found to consist of extremely minute needle-like fibres of Serpentine, 

 which sometimes stand upright, parallel, and almost in contact with each 

 other, like the fibres of asbestos (so that the film which they form has 

 been termed the 'asbestiform layer'), but which are frequently grouped 

 in converging brush-like bundles, so as to be very close to each other in 

 certain spots at the surface of the film, whilst widely separated in others. 

 Now these fibres, which are less than 1-10, 000th of an inch in diameter, 

 are the ' internal casts ' of the tubuli of tho Nummuline layer (a precise 

 parallel to them being presented in the ' internal cast ' of a recent Am- 

 phistegina in the Author's possession); and their arrangement presents 

 all the varieties which have been mentioned ( 488) as existing in the 

 shells of Operculina. Thus these delicate and beautiful siliceous fibres 

 represent those pseudopodial threads of sarcode, which originally tra- 

 versed the minutely-tubular walls of the chambers; and a precise model 

 of the most ancient animal of which we have any knowledge, notwith- 

 standing the extreme softness and tenuity of its substance, is thus pre- 

 sented to us, with a completeness that is scarcely even approached in any 

 later fossil. 



496. In the upper part of the ' decalcified ' ^specimen shown in Plate 

 xvii., fig. 2, it is to be observed that the segments are confusedly heaped 



