jMLy 1 8, 1889] 



NATURE 



269 



inside and outside, by thin " ^pidennes," which appear to 

 have been partially calcified. Aristocystis is the genus 

 affording the most abundant material for the study of 

 these coverings, but they have also been observed in 

 Arachnocysiis, Craterina, Dcndrocystis^ Neocystis, Oro- 

 cysfis, Proieocystis, and Rhombifera {i.e. R. bohemica) : of 

 Deidocystis and Pirocystis, the inner lining is the only 

 part known. In Mitrocystis the great difference between 

 the internal and external impressions {vide p. 65) may be 

 due to the presence of an inner lining. The outer layer 

 is continuous, and covers up all the pores of the body- 

 plates ; sometimes it is even thick enough to conceal 

 their sculpture and the sutures. The inner lining appears 

 to have passed up into the pore-canals, and to have been 

 itself pierced at those points. There is, of course, no 

 reason to suppose that this structure of the test was com- 

 mon to all the forms which we call Cystidea, but the 

 observations of Barrande have undoubtedly opened a 

 new field of investigation. 



It has hitherto been usual to classify the Cystidea 

 according to the disposition of their pores. Facts 

 recorded in the present work are opposed to such an 

 arrangement. In the simplest type, canals run right 

 through the inner lining and the principal layer of the 

 test, usually in a curved course : we may call these " hap- 

 lopores " ; they are best seen in Aristocystis bohemica. 

 In other species oi Aristocystis, e.g. A. subcylindrica and 

 A. grandiscuttim, the distal ends of two adjacent haplo- 

 pores are often connected by a horseshoe-shaped groove 

 on the surface of the middle layer. In the allied Cra- 

 terina this groove appears to have expanded into an 

 oval depression into which open the two canals. This, 

 the famihar diplopore, is also seen m Proteocystis. In all 

 these forms the pores do not open to the exterior, but are 

 closed by the outer layer : the horseshoe groove and the 

 oval depression are therefore nothing more than closed 

 horizontal canals connecting the vertical canals ; in these 

 genera, at all events, the canals can have had nothing to 

 do with tube feet. The structure of the hydrospire 

 elements in such forms as Echinosphaera, Arachnocysiis, 

 Caryocystis, is essentially the same ; that is to say, two 

 vertical canals connected by a closed horizontal canal : 

 the only difference is that the two vertical canals are 

 separated by a suture line instead of being on the same 

 plate. From a position originally just beneath the outer 

 layer these horizontal canals seem to have sunk into the 

 main layer of the test, and in Caryocrinus they are 

 actually internal. Rhombifera {sc. bohemica) and Homo, 

 cystis show the gradual specialization of certain portions 

 of the test as canal-bearers : here it seems that the 

 horizontal canals become more developed in proportion ; 

 in fact, they appear eventually to assume the whole 

 function, whatever it may be. In such forms as Lepado- 

 crinus they alone remain, and here they no longer con- 

 nect the adjacent plates, but are transformed into two 

 separate sets of folds. As to the function of these organs, 

 Barrande regards it as respiratory, but drops the term 

 hydrospire, for the quaint reason that they show no 

 spiral structure. Instead, he calls them hydrophores. 

 But he also applies this term to certain organs which can 

 hardly be homologous with pore-rhombs. Around an 

 opening, the interior of the calyx appears divided into 

 five compartments ; from the opening there passes into 



each compartment a branch, which subdivides into five 

 or six smaller branches, and each of these ends in a 

 double tubercle. It is impossible to see whether these 

 branches were hollow tubes or no ; they have no connec- 

 tion with the exterior, except at the central opening. 

 These hydrophores palmces occur in Aristocystis, Piro- 

 cystis, and Craterina, forms already abundantly provided 

 with canals : though simulating the hydrospires of Blasto- 

 idea as seen in section, they have an exactly contrary 

 position. Neumayr thinks that the opening which they 

 surround is the mouth, and that they are subtegminal 

 ambulacral grooves. How this can be, when their distal 

 ends are unconnected with the exterior, is not easy to 

 understand. Barrande, moreover, cannot say whether 

 they are at the oral or aboral pole. A comparison of 

 Figs. 28 and 32 on PI. XXIX. suggests that they are at 

 the aboral end, and that the large opening represents the 

 axial canal of the stem. May they not be connected 

 with nerve-cords passing from a chambered organ } 



As regards the major openings of the Cystidean calyx, 

 the accepted views are confirmed by Barrande. Some- 

 what unaccountably, however, he speaks of the anus of 

 Agelecrinus as the mouth (p. 84) ; perhaps he intended 

 to quote Vanuxem, in which case the oversight is the 

 editor's. In addition to the mouth, anus, and genital 

 pore, Barrande describes for Aristocystis a slit-like 

 aperture close to the mouth. This he regards as an organ 

 hitherto unknown. At the same time he points out that 

 its position is similar to that occupied by the peculiar 

 folded structure described by Volborth in Sphaeronis 

 Leuchtenbergi. More closely still does it resemble the 

 " reniform groove," or " semilunar pore," figured by 

 Forbes in the fossils which he called Apiocystis and 

 Echinoencrinus. We notice that Atelecystis is still 

 quoted as having anal and genital openings, although 

 neither in this genus nor in its near relation Mitrocystis 

 do the Bohemian species show them. The fact is that 

 even in A.forbesianiis they have never really been found ; 

 the appearances described are only accidental. The true 

 openings in all the Anomalocystidae must be looked for 

 in the neighbourhood of the arms. 



Although Barrande enters very largely into the question 

 of the arms, he has brought out no results of importance. 

 The arms of Arachnocysiis infaicsta are perhaps more 

 developed than those of any other cystid : it is curious 

 how closely they resemble the stem in' structure ; one 

 would almost imagine that they had originated in the 

 same manner. A similar structure is observable in the 

 process of Dendrocystis which Barrande regards as a 

 ventral tube : there is, however, no reason to call it 

 anything but an arm ; and at the same time we may 

 remember that the ventral tube of Poteriocrinus, to 

 which Barrande and Trautschold have compared it, is 

 not improbably itself derived from an arm. 



Such are the points of greatest morphological interest 

 in this fascinating work. Many peculiar and wonderful 

 forms, such as Acanthocystis, Trochocystis, Neocystis, 

 Mespilocysiis, Ascocystis, and Cystidea tiugatula, must be 

 passed by with mere mention. The question of classifica- 

 tion we have left where Barrande preferred to place it — 

 on the shelf. Systematists may even wish that many a one 

 of these new fdrms had been left in the e.z.rt\\—irrepertum 

 et sic inelius situm — and not dug up to disturb existing 



