laminated bone 

 osteogenic layer of 



periosteum 

 fibrous layer of 

 periosteum 



first generation 



Haversian system 



last generation 

 osteons 



Volkman's ca 



osteogenic lining 

 of Haversian canol 



Figure 8-24. Stereodiagrammatic piece of mammalian bone showing 

 the outer and inner laminate layers between which are several genera- 

 tions of Haversian systems. 



spaces filled in with new osteon systems. In contrast to den- 

 tine, the cells are multipolar with arborizing (branching) 

 processes extending out in every direction. Acellular bone is 

 without lacunae or channels for cell processes. 



Among primitive vertebrates Haversian or osteon systems 

 were not generally employed in bone structure; the bone 

 was of a simple laminar type, being added to at its surfaces. 

 This laminar bone was penetrated by blood vessels, the 

 focal point for tubular arrangements of cell systems. The 

 interior of thick bones was generally spongy, and the mar- 

 row spaces were irregularly crossed by thin splints (trabe- 

 cles) or plates of bone. In development, an animal may 

 start with laminar bone, erode parts of this away around 

 blood channels (through the action of osteoclasts), and re- 

 place these regions with osteon systems. The marrow spaces 

 may also be filled with osteons. In this way, an adult bone 

 may be a complex of structural types. 



In some groups, bone has been modified by the removal 

 of cells. In the teleost fishes, the bones generally are thin 

 and formed of laminae between which there are no cell 

 lacunae. 



In considering the scale structure of the various kinds of 

 vertebrates, it seems best, in order to retain the sequence of 

 simple to complex, to begin with the agnaths. 



Agnaths 



The earliest scales and plates (Middle Ordovician) are 

 assumed to be those of heterostracan agnaths, Aslraspis, 

 Pycnaspts, and Eriptychtus. These identities are based on the 

 histological differences observed in sections cut from the fos- 

 sil scales and plates. 



The plates of these genera have scattered superficial 

 tubercles formed of dentine (Figure 8-25). The plate is 

 formed of aspiden, an acellular bone. The aspiden is per- 

 meated with irregular and anastomosing channels which 

 extend up into the bases of the tubercles as pulp cavities 

 and open through the dorsal and ventral surfaces of the 

 plate. These channels were occupied in life by blood ves- 

 sels and the cells producing the aspiden. The aspiden 

 between channels is permeated by tubules for collagenous 

 Sharpey's fibers. 



In some of the heterostracans, the pteraspids, the plates 

 have a thin enamel cover for the tubercles or ridges. The 

 outer and basal plates of aspiden laminae are separated by 

 thin verticle septa. These septa form a polygonal pattern 

 when seen in frontal section. The large chambers enclosed 

 by these septa were probably filled with marrow. 



The osteostracan plate is like that of the heterostracan, 

 except that it is formed of cellular bone (Figure 8-26). The 

 dentine is cellular, that is, semidentine or mesodentine. A 

 thin layer of enamel covers the tubercles or the general 

 surface of the plates. The plate may have large chambers 

 or a spongy interior. 



The development of the plates of Tremataspis, an early 

 Upper .Silurian form, has been described. It is of interest 

 that all of the specimens (2000 or more) are of adult size 

 and that all but a few are fully armored. The conclusion is 

 that armor did not develop until the animal was fully 

 grown and once present, the animal's size was fixed. The 

 outer layers of mesodentine appeared first; then bone en- 

 closed the mucous canals of the sensory system (Chapter 

 13) and the underlying and parallel vascular channels. The 

 laminated basal part began to form. The space between the 

 superficial mesodentine layer and this base was filled in 

 gradually with spongy bone, while new laminae were added 

 to the inner surface of the base. 



It has been assumed that at least some immature osteo- 

 stracans were covered by a shagreen (referring to the rough 

 surface) of small denticles. These denticles consist of dentine 

 and semidentine (or mesodentine) with an interior pulp 

 cavity, covered externally by a layer of enamel — that is, they 

 are placoid scales. If the young were covered by such scales, 

 growth of the armor would involve connecting these and 

 adding deeper layers just as in Tremataspis. Sometimes the 

 first placoid scales are retained on the surface of the armor 

 or they may be replaced with secondary and larger denti- 

 cles. In the armor of some cephalaspids, new generations of 

 denticles are superimposed on earlier ones. As a generality, 

 later growth of the armor involved only additions to the 



SCALES 



229 



