After the examination of a large number of forms it becomes evident that the primitive 

 spicule is the terete form known as the spindle. That this is the original form is indicated by 

 two lines of evidence 1 . 



i st . The spindle is found in all of the genera and nearly all of the species of the 

 Muriceidtz, and it is the only form of spicule of which this is true. 



2 nd . In studying the development of spicules that are apparently far removed from the 

 spindle in form, we find that even the most complicated of these, such as those found in 

 Echinogorgia, Acamptogorgia and Heterogorgia, are originally simple spindles which during 

 their individual development gradually assume the surprisingly complex and often beautiful forms 

 characteristic of these genera. 



It having been determined that the spindle is the basal form, it becomes possible to 

 arrange its various modifications in different directions and thus understand fairly well the origin 

 and relations of the various generic types of spicules. 



One of the simplest of these is a slightly bent spindle, the two parts of which, below 

 and above the bend, have become differentiated from each other. The moity which is exposed 

 when in situ has become relatively smooth and pointed, like an icycle. The other portion, 

 which when in situ is embedded in the ccenenchyma, is covered with warty verrucse which in 

 some instances may be more or less profusely branched. This type is found in its most perfect 

 and characteristic development in the genus Acanthogorgia, and will therefore be called the 

 " Acanthogorgia type" in this work. (See plate XIX, fig. \a). 



A nother very typical form of spicule is produced by a process which begins in very 

 much the same way as the above. A simple straight spindle becomes gradually differentiated 

 into two parts, a distal and a proximal. The distal part, exposed when in situ, is straight, 

 thorn-like and often with a smooth surface. The proximal portion sends forth a number of 

 processes which are densely tuberculate, and more or less branched and exfoliated, often very 

 complexly so. These branched processes are projected downward and outward, and when in situ 

 are embedded in the ccenenchyma of the calyx wall. This form is characteristic of the genus 

 Echinomuricea, and will be called the " Echinomuricea type". (See plate XXI, fig. ja). 



A third form which yields a good generic type is developed as follows : A small typical 

 spindle, smooth and straight, becomes tuberculate. Later it bends as in Acanthogorgia, becoming 

 arcuate and shows a constricted band or girdle near its centre. Nodules next appear on the 

 convex surface, and these nodules enlarge, branch and protrude downward; finally producing 

 much-branched, nodulated and foliated expansions which are radiate in arrangement and are 

 embedded in the ccenenchyma when in situ. While these foliated expansions of the basal part 

 of the spicule are developing, another series of changes commences by which lamellar expansions 

 project upward from the centre of the spicule. These are more or less inflated and petal-like 

 in shape, their edges being often sinuous and dentated. 



This group of folia, when best developed, finally comes to resemble the half-opened bud 



1 The author must here acknowledge his indebteduess to one of his sludents, Miss Beth. M. Portlock, for her work in 

 studying the development of the spicules of the Alcyonaiia, and in bringing to his attention a number of points that he would otherwise 

 doubtless have missed entirely. 



