SCIENCE. 



145 



ing age, the independent arcs of pores, composed of three 

 or more pair of pores, of the Echinometradae. 



In the Echinidae proper we find in the young stages the 

 same unbroken vertical line of pores, which gradually be- 

 comes changed to the characteristic generic types. We find, 

 as in the Echinometradae, an anal system closed with a single 

 plate, and an abactinal system separating in somewhat more 

 advanced stages from the coronal plates of the test. This is 

 as yet made up of a comparatively small number of plates, 

 carrying but few large primary tubercles, with fantastically 

 shaped spines entirely out of proportion to the test, but 

 which, little by little, with the increase of the number of coro- 

 nal plates, the addition of primary tubercles, and their propor- 

 tional decrease in size, assume more and more the structure 

 of the genus to which the young belongs. The original anal 

 plate is gradually lost sight of from the increase in number 

 of the plates covering the anal system, and it is only among 

 the Temnopleuridae that this anal plate remains more or 

 less prominent in the adult. In the Salinidae, of which we 

 know as yet nothing of the development, this embryonic 

 plate remains permanently a prominent structural feature 

 of the apical system. 1 



Among the Clypeastroids the changes of form they un- 

 dergo during growth are most instructive. We have in the 

 young Fibularinae an ovoid test, a small number of coronal 

 plates surmounted by few and large primary tubercles, sup- 

 porting proportionally equally large primary radioles, sim- 

 ple rectilinear poriferous zones, no petaloid ambulacra, — 

 in fact, scarcely one of the leatures we are accustomed to 

 associate with the Clypeastroids is as yet prominently 

 developed. But rapidly, with increasing size, the number 

 of primary tubercles increases, the spines lose their dis- 

 proportionate size, the pores of the abactinal region become 

 crowded and elongate, and a rudimentary petal is formed. 

 The test becomes more flattened, the coronal plates increase 

 in number, and it would be impossible to recognize in the 

 young Echinocyamus, for instance, the adult of the Cida- 

 ris-like or Echinometra-like stages of the Sea-urchin, had 

 we not traced them step by step. Most interesting, also, is 

 it to follow the migrations of the anal system, which, to a 

 certain extent, may be said to retain the embryonic features 

 of the earlier stages of all Echinoderm embryos, in being 

 placed in more or less close proximity to the actinostome. 

 What has taken place in the growth, of the young Echino- 

 cyamus is practically repeated for all the families of Clyp- 

 eastroids : a young Echinarachnius, or Mellita, or Encope, 

 or a Clypeaster proper, resembles at first more an Echino- 

 metra than a Clypeastroid ; they all have simple poriferous 

 zones and spines and tubercles out of all proportion to the 

 size of the test. 2 



When we come to the development of the Spatangoids, 

 we find their younger stages also differing greatly from the 

 adult. Among the Nucleolidae, for instance, the young 

 stages have as yet no petals, but only simple rectilinear pori- 

 ferous zones. They are elliptical with a high test, with a 

 single large primary tubercle for each plate, and a simple 

 elliptical actinostome, without any trace of the typical bour- 

 relets and phyllodes so characteristic of this family. Very- 

 early, however, this condition of things is changed, the test 

 soon becomes more flattened, the petals begin to form as 

 they do in the Clypeastroids, and we can soon Uace the rudi- 

 ments of the peculiar bourrelets characteristic of the family, 

 accompanied by a rapid increase in the number of tubercles 

 and in that of ihe coronal plates. 



Among the Spatangidae some are remarkable in their adult 

 condition for their labiate actinostome, for the great devel- 

 opment of the petals, for the presence of fascioles surround- 

 ing certain definite areas, for the small size of the tubercles, 

 the general uniformity in the spines of the test, and the 

 specialization of their anterior and posterior regions. On 

 examining the young stages of this group of Spatangoids, 

 not one of these structural features is as yet developed. 

 The actinostome is simple, the poriferous zone has the same 



1 The young of the following genera have served as a basis for the pre- 

 ceding analysis of the embryonic stages of the Desmosticha : Cidaris, Do- 

 rocidaris, Goniocidaris, Arbacia, Porocidaris^ Strongylocentrotus, Echino- 

 metra, Echinus, Tcxopneustes, Hipponoe, remnopleurus, Temnechinus, 

 and Trigonocidaris. 



a Among the Clypeastroids I have examined the young of Echinocya- 

 mus, FibuTaria, Mellita, Laganum, Echinarachnius, Encope, Clypeaster, 

 and Echinunthus. 



simple structure from the actinostome to the apex, the 

 primary tubercles are large, few in number, surrounded by 

 spines which would more readily pass as the spines of Cida- 

 ridae than of Spatangoids. The fascioles are either very in- 

 distinctly indicated, or else the special lines have not as yet 

 made their appearance ; the ambulacral suckers of the an- 

 terior zone are as large and prominent as those of the young 

 stages of any of the regular Echini. It is only little by little, 

 with advancing age, that we begin to see signs of the spec- 

 ialization of the anterior and posterior parts of the test, that 

 we find the characteristic anal or lateral fascioles making 

 their appearance, only with increasing size that the spines 

 lose their Cidaris-like appearance, that the petals begin to 

 be formed, and that the simple actinostome develops a 

 prominent posterior lip. In the genus Hemiaster, the 

 young stages are especially interesting, as long before the 

 appearance of the petals, while the poriferous zone is still 

 simple, the total separation of the bivium and of the triviurn 

 of the ambulacral system, so characteristic of the earliest 

 Spatangoids (the Dysasteridae), is very apparent. 8 



From this rapid sketch of the changes of growth in the 

 principal families of the recent Echini we can now indicate 

 the transformations of a more general character through 

 which the groups as a whole pass. 



In the first place, while still in the Pluteus all the young 

 Echini are remarkable for the small number of coronal 

 plates, and for the absence of any separation between the 

 actinal and abactinal systems and the test proper. They all 

 further agree in the large size of the primary spines of the 

 test, whether it be the young of a Cidaris, an Arbacia, an 

 Echinus, a Clypeaster, or a Spatangoid. They all in their 

 youngest stages have simple vertical ambulacral zones ; be- 

 yond this, we find as changes characteristic of some of the 

 Desmosticha, the specialization of the actinal system from 

 the coronal plates, the formation of an anal system, the rapid 

 increase in the number of the coronal plates, with a corres- 

 ponding increase in the number of the spines and a propor- 

 tional reduction of their size, the formation of an abactinal 

 ring, and the change of the simple vertical poriferous zone 

 into one composed of independent arcs. 



In the Spatangoids and Clypeastroids we find common to 

 both groups the shifting of the anal system to its definite 

 place, the modifications of the abactinal part of the simple 

 ambulacral system in order to become petaloid, and the 

 gradual change of the elliptical ovoid test of the young lo 

 the characteristic generic test, accompanied by the rapid in- 

 crease in the number of the primary tubercles and spines. 

 Finally limited to the Spatangoids are the changes they un- 

 dergo in the transformation of the simple actinostome to a 

 labiate one, the specialization of the anterior and posterior 

 parts of the test, and the definite formation of the fascioles. 

 Comparing this embryonic development with the paleon- 

 tological one, we find a remarkable similarity in both, and 

 in a general way there seems to be a parallelism in the ap- 

 pearance of the fossil genera and the successive stages of 

 the development of the Echini as we have traced it. 



We find that the earlier regular Echini all have more or 

 less a Cidaris-like look, — that is, they are Echini with few 

 coronal plates ; large primary tubercules, with radioles of a 

 corresponding size ; that it is only somewhat later that the 

 Diademopsidae make their appearance, which, in their turn, 

 correspond within certain limits to the modifications we 

 have traced in the growth of the young Diadem atidae and 

 Arbaciadae. The separation of the actinal system from 

 the coronal plates has been effected. The poriferous zone 

 has either become undulating, or forms somewhat indefi- 

 nite open arcs ; we find in all the genera of this group a 

 larger number of coronal plates, more numerous primaries, 

 the granules of the Cidaridae replaced by secondaries and 

 miliaries, and traces of a Hemicidaris-like stage in the size 

 of the actinal ambulacral tubercules. 



Comparing in the same way the paleontological develop- 

 ment of the Echinidae proper, we find that, on the whole, 

 they agree well with the changes of growth we can still 

 follow to-day in their representatives, and that, as we ap- 

 proach nearer the present epoch, the fossil genera more 

 and more assume the structual features which we find de- 



3 For this sketch of the embryology of the Petalosticha I have examined 

 the young of Echinolampus, Echinoneus, Echinocardium, Brissopsis, Agas- 

 sizia, Spatangus, Brissus, and Hemiaster. 



