FORMS OF SPICULES. 65 



(other than spherical or approximately spherical) forms 

 which arise in the interior of cells: all other definitely- 

 shaped deposits arise on the exterior of cells, and indi- 

 vidually owe their (inherited) form as a whole (though not 

 necessarily their patterns and the like) to that of the cyto- 

 plasmic surface (mould) which produces them. As before 

 stated, there is no evidence that the forms of spicules are 

 determined by internal moulds ; on the contrary, evidence 

 exists which leads us to suppose that snch moulds cannot 

 exist. 



Another and indeed the principal feature distinguishing 

 spicules from other cell-deposits is their general nature 

 which, as all authorities agree, is closely allied to that of 

 crystals and similar bodies. The geometrical symmetry of 

 the forms of many spicules, and their physical nature, are 

 both characteristics pointing to the close affinity which exists 

 between spicules and crystals, or bodies allied to crystals, 

 such as those which I have below termed '^ crystallomorphs." 

 It has already been shown how the forms of spicules differ 

 from those of crystals, but it has not yet been shown how 

 very similar, both as regards condition of formation, physical 

 properties, and variety, symmetry and complexity of form, 

 spicules are to crystallomorphs. The one feature in Avhicli 

 spicules differ from crystal-like bodies is their, in many cases, 

 obvious adaptation of form to the architecture of the or- 

 ganism — the* feature which of course has led to the supposi- 

 tion that their forms are inherited — but as I shall show later, 

 this feature can be otherwise accounted for. The assertion 

 that, despite the facts, spicules have no affinity with crystal- 

 line bodies, but are more closely allied to utter dissimilar 



various forms of spicules, according to its position in the organism, whilst 

 quite uninfluenced by the rest of the organism (since it is uncon- 

 nected with the rest of the organism). To say that the scleroblast is influenced 

 by " heredity " at a distance is merely to assert what I assert, viz. that direct 

 ontogenetic causes determine the form of the spicule and not heredity, A 

 scleroblast in all probability no more produces a spicule-mould, in the sense 

 that a nephroblast produces a uephridium, than a cell of adipose tissue swells 

 out by heredity to produce a spherical oil-drop. 



VOL. 51, PART 1. NEW SEEIES. 5 



