138 FAMILY MILIOLIDA. 



sometimes so much compressed that their two sides meet, and the internal cavity is altogether 

 obliterated (figs. 28, 29). In this last type, of which the dimensions are usually considerably 

 less than the preceding, it is not uncommon to meet with two or more long narrow rods, 

 adherent side by side, like the frustules of some Diatoms ; and as the rods are usually smaller 

 at one extremity than at the other, their direction is radiating rather than parallel (fig. 30). 

 Their surface is marked with large pores, usually disposed with considerable regularity in a 

 somewhat alternating arrangement ; the direction of this alternation varies, however, in the 

 different forms, being sometimes transverse, sometimes longitudinal, and sometimes oblique. 

 Each pore in unworn specimens is surrounded by a broad, slightly prominent lip (fig. 28) ; 

 this, however, is very commonly lost by attrition, leaving a large circular orifice (fig. 27). 

 In some instances the chambers to which these pores lead only partially adhere to each other 

 by their external walls, "junctural interspaces" being left between them, which mark (as in 

 Badylojjora glanduhm) their essential dissociation. 



202. Internal Structure. — Under whatever form Acicularia presents itself, its structure 

 is always essentially the same, — an aggregation of separate chambers, more or less closely 

 packed together. In the cylindrical forms, the cylinder may be considered to be made up 

 of a pile of annuli resembling those of Bactj/lojjora annulm in every other essential respect 

 than this, that the chambers open on the external margin of the ring, instead of its internal 

 margin. They are usually, however, packed more closely together vertically ; and those of 

 the successive rows alternate with each other, like the cells of a honeycomb, so that there is 

 no space lost between them, as is well seen in cylindrical specimens of which the surface 

 has been so much worn away that the chambers are laid completely open (fig. 32). The 

 aspect of such specimens so strongly resembles that of the most attenuated forms of Dacty- 

 lopora cyUndracea (^ 194), that the one type might readily be mistaken for the other, if 

 regard were not had to the fact that in the latter there is a pore at the bottom of every one 

 of the funnel-shaped depressions (Plate X, fig. 20 a), which does not exist in the former. In 

 those compressed forms in which the cavity is obliterated and the two sides close completely 

 together, the inner walls of the chambers will of course meet back to back ; and the same 

 kind of alternation in the position of the chambers of the opposite surfaces may be then 

 observed, as is well known to exist in the cells of a honeycomb. In some of the narrowest 

 of the rod-like forms, there are not, even at their broader ends, more than three chambers in 

 each series, two showing themselves on one side, and only one on the other ; and the 

 alternating disposition of these gives rise to the arrangement of the pores exhibited in the 

 upper part of the composite specimen represented in fig. 30. Towards their narrower 

 extremities we find the pores arranged in only a single row, as shown in the lower part of 

 the same specimen ; the chambers of the two sides then alternating simpl}' with each other. 



203. Jffinities. — It is extremely curious to trace in this type a reversion to the same 

 fundamental plan as that on which the simplest forms of Badyhpora are constructed ; the 

 entii'e composite organism, in each case, being made up of an aggregation of separate 

 chambers. Of the additional structure which is superinduced upon this, however, in the 

 higher forms of Badylopora, no trace whatever has- yet presented itself in Acicnlariri. It is 

 important, also, to observe that in these composite organisms we have another illustration 



