104 DR. MARTIN BARRY ON FIBRE. 



out of an enlarged filament; the projecting portions of the spiral threads in the fila- 

 ment, that is, the apparent segments, becoming the teeth of that fibre. And here 

 it is important to refer to a remark made in a former page (par. 16), that the filament 

 has a tendency to become membranous at the surface, and that it contains within 

 itself the elements of other filaments : both of which qualities may be recognized, 

 as the filament is passing into the toothed fibre. 



63. From my observations on vegetable structure, to be presently referred to (par. 

 68), I venture to anticipate that the toothed fibre noticed by SCHWANN in the epi- 

 dermis of a Grass, and considered by him as corresponding to that of BREWSTER in 

 the lens, will be found to have the same mode of origin. 



Facts observed in the Structure of Blood-vessels, Mould, Woody Fibre, Hair, 



Feathers, 8$c. 



64. In examining the coats of a large blood-vessel, I had noticed the filaments of 

 one stratum to cross those of another stratum at right angles. But what was my 

 surprise, when subsequently directing my attention to the structure of the arachnoid, 

 at the remarkable display of filaments in the vessels of the pia mater (fig. 16.) ! There 

 are few parts in which the flat and compound filaments (" fibres "), so constantly men" 

 tioned in this memoir, are more easily or more distinctly seen, than here. The coats 

 of such of these vessels as are empty, or nearly so, present an inner stratum of fila- 

 ments having a longitudinal direction, and an outer filament spirally crossing these. 

 In the coats of such of these vessels as are full, the outer or spirally directed fila- 

 ment is wanting. Vessels with the same structure are met with, having many times 

 the diameter of the largest of those in fig. 16. 



65. I saw in the olfactory nerve, blood-vessels having two sets of filaments sucli as 

 those now described, as existing in the pia mater ; and their diameter so small, as to 

 admit the blood-corpuscles in only a single row. It is deserving of remark, that the 

 corpuscles observed in this row, presented indications of division into corpuscles of 

 minuter size. 



66. We thus find blood-vessels, the walls of which consist of filaments, having the 

 same structure as those filaments which the blood-corpuscle forms. In connection 

 with the spiral direction of the outer filament in these vessels (fig. 16 /3), as well in- 

 deed as with many facts recorded in this memoir, I refer to the rouleaux in which 

 the blood-corpuscles are seen in the microscope to arrange themselves, as probably 

 indicating a tendency to produce spiral filaments. To form rouleaux, corpuscle joins 

 itself to corpuscle, that is to say, ring to ring, and rings, as we have seen (par. 34), 

 pass into coils. The union of such coils, end to end, would form a spiral. But the 

 formation, by the blood-corpuscles, of these rouleaux, is no less interesting in con- 

 nection with facts recorded in a former memoir ; namely, that structures, including 

 blood-vessels, may be seen to have their origin in rows of cells derived from corpus- 

 cles of the blood. 



67. I have noticed very curious resemblances in mould, arising from the decay of 



