306 FAMILY NUMxMULINIDA 



separation of the principal chambers from their lateral extensions by the transverse septa 

 {c, (■), which are perforated close to the external side of the chambers, by the large, round, or 

 oval apertures (i, li), for the passage of the longitudinal stolons. The alar extensions are 

 themselves subdivided by similar septa, so that when they are laid open longitudinally, as at 

 d, fig. 27, we see a succession of rounded chamberlets, connected by a continuous gallery, as 

 is shown on a larger scale in fig. 27, li, V . The chamberlets of successive chambers alternate 

 with one another in position, as seen at d, fig. 27, and on a larger scale in fig. 28 ; and hence 

 arises a peculiar appearance presented by sections, by which at first 1 was considerably per- 

 plexed. In fig. 29 it will be seen that whilst at b, 1, the stolon-passages open into the 

 cavities of the chamberlets, they seem at U , V to lie between lamellae of the septa, which 

 separate to give them passage ; but it becomes obvious, from a comparison of this figure with 

 fig. 28, that whilst a section across the line a a , will traverse the open cavity {(■) of the cham- 

 berlets of one row, it will traverse the narrow gallery {!/) between the aUernating chamberlets 

 of the next row ; and thus the apparent lamellse are really two distinct septa, which here 

 close together around the connecting stolon, but which presently separate again to form a 

 chamberlet. The alar prolongations widen out greatly towards their extremities ; and their 

 secondary septa (as shown at fig. 27, ./'.,/■), instead of crossing them transversely to the axis of 

 the shell, come to lie very obliquely to it, giving rise to a variation which is at first very 

 perplexing in the appearances presented by sections dividing this part of the shell either 

 transversely or longitudinally. The appearance of divarication between two septal lamellae, 

 for the purpose of giving passage to the longitudinal stolons, is here such (fig. 29, 0', //') as very 

 strongly to suggest the idea of a canal-system ; but I am satisfied that this appearance is to 

 be explained by the alternation already described as existing between the chamberlets of one 

 row and those of the next ; so that if we could trace onwards any one of these passages, it 

 would be found immediately to open-out into a chamberlet. I have not been able to discover 

 in this type any real indication of the existence of a canal-s}^stem. 



Affinities. — It is not without hesitation that I have assigned to this type the position in 

 which I have here placed it ; Messrs. Parker and Rupert Jones having strongly expressed 

 themselves (lxxx a), in favour of its intimate relationship to AJveoliiia. The balance of 

 evidence, however, appears to me decidedly in favour of its Nummidine character ; for although 

 the tubularity of the shell cannot be certainly afiirmed, yet it is verj^ strongly indicated ; and 

 its indications arc confirmed by the Nummuline character of the aperture, as well as by the 

 very close conformity to the general structure of the Nummuline type, which is admitted by 

 Messrs. Parker and Rupert Jones themselves to be presented by the simpler forms oi FusuJIua. 

 It appears to me, therefore, that the relation of Fit-mUm to Alveolina is one of ' isomorphism' 

 only, each type representing the other in its own series ; and thus a very interesting 

 parallelism is completed between the four principal plans of conformation which are presented 

 by the higher types of the Porcellanous and the Vitreous series respectively : 



Porcellani'u . V'tlrea. 



Simple complanate spiral Pcneroplis. Opeiculiiia. 



Spiral, with the chambers subdivided into chamberlets . . Orbiculina. Heterostegina, 



Cyclical, with annuli subdivided into chamberlets . . Orbitolites. Cycloclypeus. 



Spiral with elongated axis, chambers subdivided into chamberlets Alveolina. Fusuhna. 



