144 CARBONIFEROUS AND PERMIAN FORAMINIFERA. 



than on the other. Sometimes the deposit of shell-substance is proportionally so great 

 that the animal has occupied but a small part of the whole test. 



The minute structure of the shell is far better preserved than might be expected in so 

 old a fossil, and it has many points of interest. The walls throughout are traversed 

 by a multitude of very minute tubuli. In the thinner portions (best seen in the hori- 

 zontal section, PI. XT, fig. 3), these are apparently the ordinary pseudopodial foramina ; 

 in the thicker (fig. 4), though their course is more or less sinuous, they are perhaps only 

 the prolongation of the same. Quite distinct from these, there exists a series of tubes of 

 much larger dimensions, most readily observed near the ends of a transverse section, and 

 clearly defined by a magnifying power of 200 diameters or more, as in fig. 5. I have 

 been unable to make out what purpose they serve, but the presence of two distinct 

 systems of tabulation is a noteworthy fact. A condition to some extent analogous may 

 be found in other genera of Foraminifera, in Orbulina for example, but as that genus is 

 characterized by a thin and uniform shell-wall, the two cases may have nothing in 

 common in their structural significance. 



With reference to the lamination of the shell. In the true Nummulite this is a 

 character of importance, for it arises from the prolongation of the alas of the saddle- 

 shaped chambers to the umbilicus of the test, each turn of the spire forming a fresh and 

 complete investment of the whole. In Archeediscus a tendency to a similar condition 

 exists, but developed in a much less marked degree and with no approach to uniformity. 

 A section of the test highly magnified, as in fig. G, shows the successive layers of shell 

 formed by the prolongation towards the umbilicus of the crescentiform edges of the 

 tube; but the earlier portions of the tube are nearly circular (transversely), and it is only 

 in the later stages of growth, when it becomes concavo-convex in section, that it assumes 

 this investing character. 



Whether there is any essential distinction, either in structure or function between the 

 thin shell-wall and the additional deposit which makes up the thicker portion in other 

 words, whether there is as in the Nummulite a distinct primary and secondary skeleton, 

 is a question that must still to some extent be left open. It is, nevertheless, quite 

 possible at times to trace the thin line of the primary wall, even when no difference in 

 structure is observable between it and the immediately adjacent supplementary layer. 

 In the same way, though I have not been able to identify any part of the structure as 

 referrible to a true canal-system, there are appearances that continually suggest the 

 possibility of its existence. 



The foregoing description, though incomplete in many points, is sufficient to indicate 

 the Nummuline affinities of Archtsdiscus. It differs from the typical Nummulite in its 

 less complex general structure, a coiled non-septate tube taking the place of a spiral 

 line of chambers, the tube, however, showing something of the same tendency as the 

 Nummuline chambers to bifurcate laterally. 



The difficulty of determining the structure and organization of so minute a fessi! is 



