775 
 H48 
 
 Earth ! 
 Sci. 
 
/ E I K E I E V^\ 
 
 LIBRARY 
 
 1 UNIVERSITY Of 
 V CAUFOKNIA J 
 
 EARTH 
 
 SCIENCES 
 
 LIBRARY 
 
 

FOSSIL 
 
 SPONGE SPICULES 
 
 FROM THE 
 
 UPPER CHALK, 
 
 FOUND IN THE INTERIOR OF A SINGLE FLINT-STONE FROM 
 HORSTEAD IN NORFOLK 
 
 BY 
 
 GEORGE JENNINGS HINDE F. G. s. 
 
 MUNICH 1880. 
 
 Printed by Dr. Ch. Wolf & son. 
 
SCONCES 
 
EARTH 
 
 INAlORAl-DISSEffiTION 
 
 7.VLI 
 
 Erlangung der akademischen Doctorwiirde 
 
 bci der 
 
 philosophischen Facultat 
 
 der 
 
 Ludwig - Maximilians - Universitat 
 
 MUNCHEN. 
 
PREFACE. 
 
 In the following pages I have attempted to give 
 a description of the forms and affinities of a number 
 of sponge spicules which were present, with a number 
 of other organic remains, detached and heterogenously 
 mingled together, in the cavity of a single flint stone, 
 from strata of the Upper Chalk Formation, at Hor- 
 stead, a village but a few miles distant from the city 
 of Norwich, England. 
 
 The way in which I met with the material which 
 yielded these sponge spicules is as follows: Returning 
 to my native city after an absence of several years 
 during which I had studied the Palaeozoic rocks of 
 North America, my interest was naturally excited by 
 the contrast between the chalk and the limestones of 
 the older rocks, with which I had become familiar, and 
 I determined to examine the numerous pits, as they 
 are termed, in which sections of the chalk were ex- 
 posed to view, in order to become farther acquainted 
 with the strata and to collect fossils for future study. 
 With this purpose I went one day to a pit at 
 Horstead, but found on my arrival that some long time 
 had elapsed since the chalk had been worked ; for al- 
 ready the section was partly obscured by a talus of 
 gravel which had fallen from the overlying beds of 
 rolled and stratified flints, and by the crumbling of 
 the chalk itself, so that the prospect of obtaining fos- 
 
sils did not seem very promising. A search among 
 the debris yielded only two or three specimens of 
 Terebratula carnea, a few more or less perfect ex- 
 amples of Ananchytes ovatus, and the very common 
 Belemnitella mucronata which could be seen every- 
 where in the face of the section. I began to think 
 that nothing more than these common everyday fos- 
 sils of the chalk would be the result of my search, 
 and was about quitting the quarry when I thought 
 it worth while to observe more particularly the 
 large variously shaped nodules of flint which had 
 been left, strewn over the floor of the pit, after the 
 removal of the soft chalk in which they had been im- 
 bedded. One of the larger of these flints, about 
 a foot in diameter, had been splintered, apparently 
 by the sharp frosts of the preceding winter, and pre- 
 sented, instead of a solid mass of stone as is usual 
 with the majority of the nodules, a central cavity, 
 which contained a quantity of material resembling fine 
 flour in appearance and feel, and of a creamy-white 
 tint. An examination with a hand-lens showed that 
 this floury material abounded with minute fossils, more 
 particularly sponge spicules. Feeling at once certain 
 that I had fallen in with a rich prize, I spread out a 
 newspaper which I had brought to wrap up my fossils, 
 and carefully emptied into it the contents of the cavity, 
 which may have weighed when moist about a couple 
 of pounds. The cavity appeared to have been com- 
 pletely inclosed by the flint, which had thus hermeti- 
 cally sealed up in its interior and preserved unharmed 
 from mechanical injury, a small portion of the mud of 
 the Cretaceous ocean. In fact, this material pro- 
 mised to afford as good evidence of the kind and 
 
variety of the ocean life of that period, as the mate- 
 rial dredged up from the bottom of the present seas, 
 did of the existing fauna. 
 
 It was not until some time after that I had dis- 
 covered this material , that I became aware that Mr. 
 Joseph Wright, F. G. S. of Belfast had also, a few 
 years previously, ascertained, that many of the flints 
 in the chalk of the North of Ireland, contained depo- 
 sits, similarly rich in fossils, and that he had published 
 a list of those which he had discovered in the con- 
 tents of these stones. As, however, it did not ap- 
 pear that these microzoa had been previously noticed 
 in the chalk Flints of the East of England, a locality 
 several hundreds of miles distant, and in strata now 
 entirely discontinuous from the beds of the Irish chalk, 
 I determined to work out the contents of my disco- 
 very for the sake of seeing how far the fossils from 
 these two places agreed with each other and with 
 those of the cretaceous strata of Germany and else- 
 where. Further I hoped to be able to show the 
 great variety of the various forms of life which in 
 such a small quantity of material and from a single 
 locality were mingled together in the mud of the Cre- 
 taceous ocean, and consequently I resolved to limit 
 my investigation to the contents of this single hollow 
 flint. 
 
 I prepared this flint meal, if thus it may be term- 
 ed, by carefully washing away the finer particles; 
 which reduced it when dry to about 3 or 4 ounces in 
 weight. By this means the fossils were brought in- 
 to a small compass, but even so, it involved pretty 
 constant work for several weeks to look over and pick 
 out, under a strong simple lens, the principal fossils 
 
from this residue ; for every spicule, foramenifer and 
 entomostracan shell had to be separated from the mass 
 by means of a moistened hair and afterwards these 
 specimens had to be classified and then mounted for 
 microscopic examination. I found the larger part of 
 the various forms in the first fortnight of working at 
 the material, but as nearly every day after this my 
 search was recompensed by finding one or two new 
 objects, I continued at the work until I reasonably 
 thought that I had obtained specimens of every form 
 present in the deposit. The result far exceeded my 
 anticipations. The variety of fossil forms, all micro- 
 scopic in point of size, in this mere handful of mate- 
 rial, and their beautiful state of preservation proved 
 extraordinary. There were fossilized remains , entire 
 or fragmentary, of Foramenifera, Sponges, Echinoderms, 
 Annelids ; also of Cirripedia, Ostracoda, Polyzoa, Bra- 
 chiopoda, Lamellibranchiata and lastly, Fishes; but by 
 far the larger part of the fossils, belonged to the fo- 
 ramenifera, sponges, and ostracoda. 
 
 It was a subject for consideration as to which of these 
 three different groups of fossils should be first studied ; 
 but as it appeared that less was known of the sponge 
 remains of the English Chalk than of the foraminifera 
 and ostracoda of this formation, I made up my mind 
 to investigate first the fossils of this class. In order 
 however to be able satisfactorily to ascertain the affi- 
 nities of these isolated spicules and morsels of sponge 
 skeleton, I found that it would be necessary to com- 
 pare them with the more or less perfect fossil sponges 
 whose structure had been already determined and that 
 the only collection of sponges which would afford this 
 opportunity for comparison was that made by Pro- 
 
. 7 -. 
 
 fessor Dr. Zittel of the Royal Palaeontological Museum 
 of Munich. - - I therefore resolved to study my fos- 
 sil spicules at Munich, and on my arrival in that city 
 I obtained from Prof. Zittel not only the freest oppor- 
 tunities for examining the rich collection of sponges 
 in the museum, as well as the numerous microscopical 
 preparations which had been made to illustrate their 
 minute structure, but at the same time received from 
 him the most cordial support and assistance in the pro- 
 gress of my work, for which I desire to express my heart- 
 iest thanks. I am further indebted to Prof. Zittel for 
 the gift of a small quantity of material, rich in sponge 
 spicules, from the upper Cretaceous Formation of Coes- 
 feld in Westphalia which yielded me a great variety 
 of forms of great value, from their beautiful state of pre- 
 servation, for comparison with those from Horstead. 
 
 The time involved in describing these sponge spic- 
 ules has been greatly prolonged by the necessity of 
 making micrometer measurements of all the forms, not 
 only in order to be able to make comparisons of their 
 dimensions with those of spicules described in other 
 works, but also to be able to state the real size of 
 these fossils and thus allow a true idea to be formed 
 respecting them. I have stated the measurements 
 in millimetres and decimal parts of the same, so as to 
 avoid giving to those students of sponges who make 
 use of this scientific scale, the inconvenient task, 
 which I have myself experienced, of reducing into mil- 
 limetres the fractional or decimal parts of the English 
 inch, in which the measurements of these small objects 
 have been expressed by most English authors who 
 have described them. 
 
 For the faithful representations of these spicules, 
 
which are given in the accompanying- plates, I am in- 
 debted to Herr Conrad Schwager, Assistant in the 
 Palaeontological Museum at Munich, who, at very short 
 notice, kindly undertook the task of drawing and li- 
 thographing them. As far as possible the figures 
 have been drawn to the same scale of 20 diameters, 
 in order to enable theirhrelative dimensions to be ap- 
 preciated. In a few instances, in which a larger 
 scale has had to be adopted, a representation of the 
 object is also given, where practicable, on the same 
 scale as the other spicules. In comparing the fig- 
 ures given in these plates with those of similar spicules 
 which are given in the beautiful illustrations of Prof. 
 Zittel's work on Coeloptychium and in Mr. Carter's fig- 
 ures of the spicules from the Haldon Green Sand, it is 
 very necessary to bear in mind the different scale to 
 which these figures have been drawn. In Professor 
 Zittel's work the detached spicules on the Plates IY-VII 
 are represented one hundred times magnified, and thus 
 five times larger than in my Plates, and the Haldon 
 spicules are drawn to the scale of seventy-five diame- 
 ters or three and a half times larger. Bearing this 
 fact in mind it will be noticed that most of the spi- 
 cules from the Horstead chalk are actually larger than 
 those from Germany and the South- West of England. 
 I may add that the accuracy of the drawings of the 
 Horstead spicules is further assured by their having 
 been done by means of the Camera lucida. 
 
 Palaeontological Museum Munich. 
 
 November 1880. 
 
FOSSIL SPONGE SPICULES 
 
 FROM THE 
 
 UPPER CHALK. 
 
 FOUND IN THE INTERIOR OF A SINGLE FLINT-STONE FROM 
 HORSTEAD, NORFOLK. 
 
 BY 
 
 GEORGE JENNINGS HINDE. F. G. s. 
 
 The sponge spicules herein described were obtained from 
 the material contained in the interior of a nodular flint, about 
 one foot in diameter, which had been imbedded in the strata 
 of the Upper Chalk Formation at Horstead, near Norwich. - 
 In the Chalk pits or quarries in this neighbourhood not only 
 are the nodular flints arranged in horizontal layers coinciding 
 wich the stratification of the Chalk but they are also disposed 
 in vertical layers, and the nodules thus arranged vertically are 
 often of much larger size as well as different in form to those 
 of the horizontal beds. Many of these larger flints are more 
 or less cylindrical and either hollow throughout like a tube or 
 with a deep cup-shaped depression at one end which has 
 led them to be compared with the large horny sponges called 
 Neptune's cups, though there is no evidence to show that they 
 are more allied to these sponges, beyond a rude resemblance 
 in outer form, than the ordinary shapeless nodules of the hori- 
 zontal layers. On account of their peculiar form, these cy- 
 lindrical, partly hollow masses of flint have been termed pot- 
 stones or paramoudras. They have by no means a general 
 distribution in the upper chalk strata, though they are of fre- 
 quent occurrence at Horstead and are also met with in many 
 of the Chalk pits near Norwich, and they appear as well in 
 
10 
 
 the same geological formation in the North of Ireland. These 
 paramoudras are more frequently hollow than the ordinary 
 nodular flints though they do not uniformly contain the fine 
 powdery material which was present in the specimen I met 
 with, but occasionally, as noted by Sir Charles Lyell, they 
 have an internal cylindrical nucleus of pure chalk much harder 
 than the ordinary surrounding chalk and not crumbling to 
 pieces like it when exposed to the winters frost. In com- 
 mon with the other chalk flints of this district the exterior 
 surface of the paramoudras is white and rough, while the 
 interior is of a dark tint. 
 
 The particular specimen which yielded me the fossiliferous 
 powder was about a foot in diameter and more spheroidal than 
 the generality of the potstones. The manner in which the 
 interior cavity became inclosed may be understood by supposing 
 that the process of the deposition of the silica , by which 
 the ordinary cup-shaped paramoudras were formed , was con- 
 tinued further, until the cup was arched over with a solid 
 layer of flint, and thus the soft incoherent ooze included within 
 was preserved safely from the effects of all further mechanical 
 influence. Without entering here into the vexed question 
 about the formation of the flints in the chalk, it may be as 
 well to notice that the beautifully perfect state of preservation 
 of the various delicate fossil organisms in the interior of this 
 flint , when compared with the nearly complete obliteration of 
 their structures in the enveloping chalk points to the conclusion, 
 that the period in which the flints were formed must have been 
 previous to that consolidation of the mass of the chalk by which 
 the smaller fossils were mostly destroyed. The flint powder 
 or flint meal which was preserved in the cavity of this flint, 
 like the ashes of the dead in a funereal urn , had the appear- 
 ance of a very fine flour of a creamy yellow tint, thus differing 
 but slightly from the soft earthy white chalk in which the 
 flints are here imbedded. The finest portions of the flint- 
 meal, under high powers of the microscope were seen to be 
 
. T T ^ 
 
 composed either of amorphous particles or fragments with a 
 rounded outline, probably portions of the chambers of foramenif- 
 era. Besides these, there were a few straight rod-like par- 
 ticles, which may have been derived from sponge spicules but 
 I could not detect any indications of the presence of cocco- 
 liths. After washing away the finer particles , the residue 
 consisted, almost exclusively, of the entire and fragmentary 
 skeletons of foramenifera , sponges and entomostraca, with a 
 scanty intermixture of the remains of other organisms. There 
 were but very few free, minute, mineral particles in the deposit 
 and these, so far as I could judge, were iron pyrites. On 
 subjecting a portion of the material to the action of Nitric 
 acid, and heating; there was at first a certain amount of ef- 
 fervescence, but the result proved that only a small portion of 
 the material was dissolved by the acid, and that the greater 
 part, including fossil shells, which, like those of the forameni- 
 fera, were calcareous in their original structure, had now be- 
 come either wholly or in part silicified. The material itself, 
 previously of a creamy yellow tint became, after treatment 
 with the acid , of a snowy whiteness. Being desirous of as- 
 certaining the extent to which the different fossils were affected 
 by the acid , I tried examples of different forms separately 
 with the following results ; 
 
 ri Sponge spicules. These were not perceptibly affected 
 by the acid and retained the same optical appearances 
 both before and after treatment. 
 
 2" d Foramenifera. The shells, previously opaque, or 
 merely translucent by transmitted light , and white by 
 reflected light, became as transparent as glass and the 
 walls appeared reduced to great thinness. 
 3 r Entomostraca. A somewhat similar change was effec- 
 ted in the shells of these animals as in those of the 
 foramenifera, but the residual siliceous shell was thicker 
 and less transparent. 
 4 l Echinoderms. Small fragments of shell of these orga- 
 
12 
 
 nisms were nearly entirely dissolved by the acid leaving 
 only a slight amorphous residue. 
 
 5'!? Mollusca. Detached prisms of the shells of Inoceramus 
 which were very numerous in the deposit, were strongly 
 attacked by the acid and completely dissolved , but 
 very minute shells of a species ot Ostraea though at- 
 tacked by the acid yet retained the form of the shell 
 perfect in silica. 
 6 l * Fish-scales. These were slowly but completely dis 
 
 solved. 
 
 7 1 ' 1 Coprolites. Small cylindrical pellets most probably 
 coprolites, were readily attacked by the acid and dis- 
 solved with the exception of a residual reddish floc- 
 culent sediment. 
 
 It will thus be seen that the degree, in which the various 
 fossils were silicified and suffered alteration in their originally 
 calcareous structure, varied considerably in amount ; that whilst 
 the foramenifera and entomostraca became so infiltrated with 
 silica as to retain the perfect form ol their shells in this mine- 
 ral, others, like the prisms of Inoceramus, though exposed to 
 the same influences , had completely retained their normal 
 condition. 
 
 Structure of the Sponge Spicules. 
 
 The Sponge spicules of this deposit, including under this 
 term the individual siliceous bodies of exceedingly varied forms, 
 which, either disconnected or attached together, form the ske- 
 letal support of the living organism, are exceedingly numerous, 
 and with the exception of a few small fragments of the ske- 
 leton net-work of the Lithistidae and Hexactinellidae are isolated 
 from each other. Seen under a low power and by reflected 
 light the spicules with few exceptions have a complete outline 
 and an aspect like that of ground glass. When mounted in 
 Canada Balsam and examined by means of a higher power 
 
. ,f 2 __ 
 
 with transmitted light, the surface of the spicules appears to 
 be rough, and pitted all over with small depressions, which 
 in some examples are so strongly marked as to give a retic- 
 ulated aspect to the surface. In the larger spicules the erosion 
 which has produced the pitted surface has not affected the 
 form of the spiculc, but in the smaller and more delicate spi- 
 cules , it has almost eaten through the spicule and rendered 
 it, particularly near the extremities, very rough and jagged. 
 As will be readily understood, the delicate spines with which 
 the spicules of both fossil and existing sponges are frequently 
 adorned, would be the first to be destroyed by the erosive 
 action and still readier would the minute flesh-spicules be dis- 
 solved by it, so that no expectation could be entertained ot 
 discovering these microscopic bodies in the material. The 
 interior of the spicules shows numerous minute ill-defined 
 brownish spots from which , as from so many centres, ex- 
 tremely small radial fibres extend in all directions. When 
 examined by polarised light these fibrous rays display pris- 
 matic tints similar to those peculiar to chalcedony, thus showing 
 that the silica in these spicules, has been changed from the 
 colloidal condition of the mineral to the crypto-crystalline state 
 of chalcedony. In the brownish spots in these spicules the 
 prismatic tints are far less clearly displayed than in the fibrous 
 portions. There arc however other spicules in the material, 
 belonging to various genera of sponges, which, instead of the 
 dull glassy aspect, common to the great majority of these 
 sponge remains, exhibit, by reflected light, a white snowy ap- 
 pearance, and when these snowy-white spicules are mounted 
 in Canada balsam they display a thin outer coating of silica, 
 which is transparent and also gives prismatic tints in the same 
 manner as the ordinary spicules; within this, however, the 
 silica is of a brownish granular aspect as if intermingled with 
 ochreous particles, and does not give the prismatic tints under 
 polarised light. The boundary between the brownish, ap- 
 parently amorphous, silica in these spicules and the outer layer 
 
14 
 
 of Chalcedonic silica is very irregular, the one form apparently 
 gradually merges into the other. It would seem that the 
 silica in these examples, to a certain extent retained the cha- 
 racters of the originally colloidal silica of the spicules, whilst 
 in the majority of the spicules in this material the metamor- 
 phosis of the silica had been extended further , until , with 
 the exception of a few minute spots, the whole of the spicule 
 had become changed into chalcedony. I may remark here, 
 that so distinctly are 'the sponge spicules in this deposit shown 
 by polarized light, that, in a small quantity of the flint-meal 
 mounted in balsam , the minutest particle of spicule at once 
 reveals itself by the prismatic tints which it exhibits, and I 
 availed myself of this fact to determine the characters of some 
 delicate lamina, which otherwise had the appearance of small 
 portions of shell. 
 
 Not only has chalcedonic silica thus replaced the original 
 colloidal silica in these spicules, but it has also, in the major- 
 ity of instances, completely filled their interior canals, so that 
 the spicules now appear solid throughout. Here and there 
 however, spicules are met with , in which the canals are either 
 wholly or in part preserved , but even in these instances they 
 do not present the form of regular even tubes as in recent 
 sponge spicules, but are abnormally and irregularly enlarged. 
 Sometimes the canal yet extends throughout the length of the 
 spicule, in other instances only a small part in the upper and 
 thicker part of the spicule is to be seen. The canal, where 
 retained , appears to be hollow , or infiltrated more or 
 less with ochreous particles, and an example of this latter 
 in which the canal is shown as a thin black tube extending 
 along the central axis of the spicule is given on Plate V. 
 
 % 30- 
 
 The present condition of the sponge spicules in the interior 
 of this Horstead flint, thus shows a great alteration from their 
 original condition. The smooth surfaces have become rough 
 and eroded, the delicate spines have been dissolved away, and 
 
the amorphous silica is, for the most part, changed into crypto- 
 crystalline chalcedony, which has also filled up and obliterated 
 the interior canals. They thus furnish corroborative evidence 
 of the instability of silica in its amorphous condition, and show 
 that even where sponge spicules ,have been preserved tolerably 
 complete in outer form yet their interior structure has been 
 considerably modified. A similar change has also been 
 noticed in the mineral condition of sponge spicules from other 
 formations and localities. Thus Mr. Carter (An. and Mag. 
 Nat. Hist. S. 4, Vol. 7, p. 112 et seq.) describes the spicules 
 from the upper Greensand of Haldon near Exeter as being 
 changed into chalcedony and having their surfaces eroded 
 and the interior canals partly or wholly destroyed. The 
 condition of the spicules from the flints of the North of Ire- 
 land, discovered by Mr. Wright, is not stated (List, of the Cre- 
 taceous Microzoa of the North of Ireland), but as the calca- 
 reous remains in these flints were infiltrated with silica, like 
 those from Horstead , it may be inferred that the sponge spic- 
 ules had been altered in a similar manner. I find too, that 
 a similar alteration in the mineral condition of the silica has 
 taken place in sponge spicules forming thin beds in the Lower 
 Green Sand (Neocomian) in Surrey, which I have lately dis- 
 covered. This change, however, in the condition of spicules, 
 occurring, as in the instances mentioned, detached and scattered 
 in the interior of flints or in beds of Green sand, does not 
 appear to have taken place invariably, and a noted exception 
 is seen in spicules found in connection with Coeloptychium in 
 the upper Chalk Strata of Coesfeld in Westphalia. These 
 have been described and figured by Professor Zittel in his 
 work on Coeloptychium (Abh. d. k. Ak. d. Wiss. XII Bd., 
 III. Abth.) and present a great contrast to the spicules from 
 Horstead. The Coesfeld spicules have, by reflected light, 
 a smooth , white porcellanic aspect , and when mounted in 
 Canada balsam, their outlines under the microscope are still 
 seen to be smooth, and the interior canals are equally as per- 
 
i6 
 
 feet as in recent spicules. Another feature of the little 
 change which these spicules have suffered, is shown by their 
 being neutral to polarised light, thus proving that the silica 
 still retains its colloidal condition, although it has lost that 
 beautiful clear glassy appearance which characterizes the spic- 
 ules of recent sponges. The state of preservation in which 
 these Coesfeld spicules are found may, however, be regarded 
 as exceptional, in the majority of instances in which detached 
 spicules have been noticed the silica appears to be now in the 
 form of Chalcedony. 
 
 As I have already mentioned, the eroded condition of 
 the larger spicules plainly indicates that none of those much 
 smaller forms which have been termed flesh-spicules , have 
 stood any chance of being preserved, and therefore, for the pur 
 poses of classification, the characters of the skeleton spicules 
 will alone have to be relied upon. Of these I separated 
 some thousands of specimens and they present the most ma- 
 nifold varieties both of form and size. In dimensions they 
 vary between o.iSmm., the diameter of small globo-stellates, 
 and the comparatively great dimensions of 8.77 mm., the length 
 of the longest trifid spicules. The various forms comprise straight 
 and curved cylindrical acerates, conical and fusiform acuates, a 
 great variety of trifid spicules (i. e. those with an elongated shaft 
 and a head of three rays) ; qdadrifid spicules , globates and 
 stello-globates, contorted skeleton spicules of Lithistids and 
 free and combined six-rayed spicules. Besides these different 
 forms of isolated spicules there are a few specimens in which 
 small fragments of the mesh-work of the sponge, showing the 
 spicules naturally attached together, have been preserved. A 
 glance at the figures given in the different plates will better 
 enable an idea to be formed of the great diversity presented 
 by these spicules , mixed up together in the interior oi 
 this single flint, than can be conveyed by a mere verbal de- 
 scription. 
 
In endeavouring to classify these, for the most part, iso- 
 lated spicules, I have compared them with the valuable collection 
 both of recent and fossil sponges contained in the Palaeontological 
 Museum at Munich, (which was freely placed at my disposal 
 for this purpose by Professor Zittel) and with the figures and 
 descriptions of sponge spicules given by Carter, Zittel, Oscar 
 Schmidt, Sollas, Bowerbank and other authors who have de- 
 voted attention to the microscopic structure of the sponge 
 skeleton. The multitude of works and descriptions of both 
 earlier and later writers on fossil sponges, which are mainly 
 based on the outer form of the organism, leaving undescribed 
 the character of the constituent spicules, were of no value to ' 
 me whatever, since my material only consisted of the small 
 bodies by whose combination the sponge itself is formed. 
 It has been only within the last few years that the fundamen- 
 tal structure of sponges has been studied and the true cha- 
 racter of the spicular skeleton recognized and made the basis 
 of classification, and though very much remains to be done 
 in this field of investigation, the knowledge which has been 
 already obtained of the spicular structure of fossil sponges, 
 mainly through the researches of Zittel, Carter, and Sollas, 
 has made it possible in many instances, to determine the genus 
 of sponges to which these isolated and scattered spicules be- 
 long. There are however several difficulties, in connection 
 with the determination of the relationship of these detached 
 spicules, which only allow an approximate conclusion to be 
 made as to their affinities. One cause of this uncertainty 
 arises from the fact that in the skeletons of both fossil and 
 recent sponges, very many spicules get intermingled with the 
 sponge, which have only found their way in the structure by 
 mechanical means and are not integral portions of the organ- 
 ism , though they have frequently been mistaken for such, 
 and have given rise to the belief that in the same sponge 
 there could exist the most varied collection of spicules. 
 Another difficulty is owing to the fact of the same form of 
 
18 
 
 spicule being common to several different genera of sponges 
 so that it is quite impracticable to determine to which or 
 how many of these genera the form may have belonged. 
 In other sponges, there are six or seven different forms of 
 spicule in the same species, of which one or two of the forms 
 may be characteristic of the species, whilst the other spicules 
 may be common to other species as well, so that, although 
 all the different forms of spicule may be present in the ma- 
 terial, there is no certainty that they have been derived from 
 the same genus or species of sponge. The peculiar form 
 and mode of attachment of the spicules of the Lithistid and 
 Hexactinellid Sponges permits of a better identification of their 
 spicules than in those of the Tetractinellid sponges, and in 
 several instances the correspondence in form and size of the 
 spicules is so close to that of sponges already determined, 
 that no doubt can arise of their belonging to the same species. 
 Under these circumstances I have arranged these spicules under 
 the different genera with which they seemed to have the 
 closest relationship, and in only a few exceptional cases in 
 which the peculiar form or dimensions of the spicule rendered 
 it highly probable that it belonged to some hitherto unre- 
 cognized sponge , have I ventured to give a name to it , to 
 facilitate reference in the future. 
 
 Of the works which hav^ already appeared on detached 
 fossil sponge spicules to which I shall frequently have occasion 
 to refer, the first is that of Mr. H. J. Carter, F. R. S. On 
 Fossil Sponge Spicules of the Greensand compared with those 
 of existing Species. (Annals and Mag. Nat. Hist. 1871, S. 4 
 Vol. 7, p. 112, Plates VII X.) In this article, Mr. Carter 
 gives an excellent description and figures of 76 different forms 
 of detached spicules which were discovered in strata of the 
 Greensand Formation (Cenomanien) at Haldon near Exeter, 
 and shows their close relationship to existing sponges. 
 Some of these Greensand spicules are identical in form with 
 those from the Horstead flint and all of them are closely 
 
I 9 
 
 related, but there is not the variety of Lithistid and Hexacti- 
 nellid spicules as at Horstead. 
 
 In 1873-4 Mr. Joseph Wright F. G. S. of Belfast gave il- 
 lustrations of 34 different forms of spicules which he had dis- 
 covered in the interior of flints from the upper Chalk Formation 
 in the North of Ireland. (A List, of the Cretaceous Microzoa 
 of the North of Ireland. Belfast. Nat. Hist. Field-Club Report). 
 The late Dr. Bowerbank drew up a report on these spicules 
 but limited himself to giving them some very long names 
 and indicating the resemblances of a few of them to existing 
 genera. With hardly an exception, all these Irish Chalk 
 specimens are present in the Horstead flint, but the sponges 
 do not seem to be anything like so well represented in the 
 Irish as in the Norfolk Chalk, though the material examined 
 by Mr. Wright was collected from several localities, and ap- 
 pears to have been carefully searched. 
 
 Also in 1874, M. A. Rutot described and figured nume- 
 rous forms of sponge spicules from Tertiary sands of the Etage 
 Bruxellien in the neighbourhood of Brussels (Annales de la 
 Societe Malacologique de Belgique. Tome IX, 1874). M. Ru- 
 tot believed that these spicules belonged to two sponges 
 merely, though there is no doubt that several different genera 
 are represented. A few of these Tertiary spicules are simi- 
 lar to those from the Horstead flint but so far as can be 
 judged from the figures the majority of them are different 
 forms. 
 
 In 1876 Professor Zittel figured numerous spicules which 
 were from the Upper Chalk of Coesfeld and other places in 
 Westphalia. (Ueber Coeloptychium. Abh. d. k. Ak. d. Wiss. 
 XII. Bd. III. Abth.) From finding many of these spicules in 
 close connection with the skeleton of Coeloptychium, Professor 
 Zittel thought, previous to his later researches, that they might 
 have belonged to that genus. With but few exceptions, 
 the forms figured by Professor Zittel are found also in the 
 
 2* 
 
2O 
 
 Horstead flint, and they are in a much more perfect state of 
 preservation. 
 
 In describing these Horstead spicules, I propose to begin 
 first with the simple uniaxial forms and then pass on to the 
 quadriaxial spicules and the more complicated lithistid and 
 hexactinellid forms ; following the systematic arrangement which 
 is given by Professor Zittel in his Beitrage zur Systematik der 
 Fossilen Spongien. 
 
 Monactinellidae , Zittel. 
 
 To this order belong those sponges whose skeleton con- 
 sist only of those spicules which possess a single unbranched 
 interior canal. But as simple monaxial spicules are also 
 common to other orders of sponges as well, the question of 
 determining the proper relationship of these spicules, when 
 detached, as in the specimens under consideration, is some- 
 what involved. As the only method available, I have com- 
 pared the simple uniaxial spicules in this deposit with those 
 which are present in recent and fossil sponges of the order, 
 as well as with the uniaxial spicules which are found in con- 
 nection with the distinctive spicules of Geodia and other genera, 
 and I have come to the conclusion that the greater number 
 of the uniaxial spicules in this deposit more probably belong 
 to Geodia and thus formed part of the Skeleton Spicules of 
 Tetractinellid sponges , under which I shall include them. 
 Those which appear to have belonged to Monactinellid sponges 
 are not very numerous. 
 
 Acuate Spicules. 
 
 (Plate I, figs. 1015.) 
 
 Simple acuate spicules, somewhat resembling pins, but 
 with less prominent heads, curved , thickest at one extremity 
 and gradually tapering to the other, which in some instances 
 is obtusely pointed. Originally smooth, though now with 
 
21 ' 
 
 pitted surfaces. Canals obliterated. In one instance (fig. n) 
 there is a sharp bend in the upper portion of the spicule 
 whilst the lower part is straight and pointed. The largest of 
 these spicules is 2,9 mm. long by 0,13 mm. wide; the smaller 
 form (fig. 12} is 0,94 mm. long by 0,067 mm - wide. In fig. 
 1 3 , the head is much larger in proportion to the length ; this 
 spicule is 0,832 mm. long by 0,15 mm. wide. All the 
 forms are rare. 
 
 Acuate spicules similar in form but usually so much 
 smaller as to preclude a fair comparison are frequent in recent 
 sponges from the Atlantic and have been placed by Mr. Carter 
 under the genus Halichondria (Ann. and Mag. Nat. Hist. S. 4, 
 Vol. XIV 1874, p. 1 6. PI. XV, fig. 40 b, 41 b. Vol. 18, 
 PI. XV. fig. 29). Oscar Schmidt also figures a similar form 
 in Plocamia gymnagusa (Atlant. Spong. p. 62, Taf. 4, fig. I7b) 
 and Sollas has also described like spicules in other species 
 of the same genus (Ann. and Mag. Nat. Hist. July 1879 
 p. 44, PL VI, fig. 3). Acuate spicules but with more deve- 
 loped heads appear in the Haldon Green Sand (Ann. and 
 Mag. S. 4, Vol. VII, PL X, fig. 77) and also from the upper 
 Chalk at Coesfeld (Ueber Coelop. p. 37, Taf. V, fig. I 10). 
 
 There are two other forms of acuate spicules, in which 
 the thickest portion of the spicule is just below the head and 
 the spicule itself is straight and pointed (Plate I, figs. 14, 15). 
 The larger is 1,23 mm. long by 0,112 mm. wide, and 
 resembles in form a spicule from Haldon (Ann. Vol. 7, PL X, 
 fig. 78) and the smaller is 0,517 m.m. long by 0,56 mm. 
 wide. 
 
 Reniera, sp. 
 
 (Plate I, figs. 18, 19, 22). 
 
 Spicules of a regular conical form, the thicker end evenly 
 rounded and from this tapering either to a point or more fre- 
 quently ending obtusely. In rare instances, the smaller end 
 expands slightly, and the spicule becomes intermediate between 
 
22 
 
 a cone and a cylinder. Outer surface originally smooth. The 
 interior canal is frequently preserved, it is of a conical form 
 corresponding to the shape of the spicule ; in some instances 
 it extends but a short distance from the head, in others it is 
 prolonged nearly to the point of the spicule. These conical 
 spicules are very abundant and very regular in form. They 
 vary in length between 0,495 mm. and 0,832 mm. and in 
 width of the head or thickest portion between 0,135 mm. 
 and 0,225 mm. 
 
 I have also found a single specimen of a spicule nearly 
 alike in form to these small cones but characterized by the 
 surface being completely covered with small blunt spines 
 (fig. 22). No canal can be distinguished. The specimen is 
 0,45 mm. long and 0,157 mm. wide. 
 
 The only instance of spicules of this conical form, which 
 I have met with, is given by Mr. Carter as a transitory stage 
 of the dermal globular crystalloids of Pachymatisma JoJm- 
 stonia, Bowerbank (Ann. Mag. Nat. Hist. S. 4, Vol. 4, p. 9, 
 PL II, fig. 1 8 c). The resemblance is however merely in the 
 outer form, for Mr. Carter states as regards these bodies, that 
 on no occasion has he been able to detect a central cavity in 
 any stage of their development, whereas the interior canal is 
 clearly exhibited in the Horstead spicules. As regards their 
 respective dimensions I am unable to make a comparison, for 
 the scale to which Mr. Carter's figure has been drawn, is not 
 stated. These conical spicules appear to be related to the 
 cylindrical spicules present in some species of Reniera and I 
 have therefore placed them provisionally under this genus. 
 
 These conical spicules have been found, though rarely, 
 in the North of Ireland (Wright; op. cit. p. 88, PL II, fig. 3) 
 and I have also discovered them in material from Coesfeld, 
 given me by Prof. Zittel. The spinous form also is pre- 
 sent in the Upper Chalk Strata of Coesfeld. 
 
Reniera, sp. 
 
 (Plate I, figs. 16, 17). 
 
 Spicules of a cylindrical form, curved, uniform in calibre 
 throughout and rounded at both ends. Exterior surface ori- 
 ginally smooth. Canals obliterated, but in similar spicules 
 from the Westphalian Chalk a simple cylindrical canal traver- 
 ses the spicule. There are two well marked forms of these 
 spicules; the one (fig. 16), longer and proportionally much 
 thinner than the other, has a length of 0.5 1 7 mm. by 0,056 mm. 
 The shorter (fig. 17) form, in some specimens almost kidney 
 shaped, is 0,33 mm. long by 0,112 mm. wide. Specimens of 
 both forms are not uncommon. 
 
 Spicules of existing sponges, nearly approaching in form 
 to these, have been discovered by Mr. Carter in species of Re- 
 niera from the Gulf of Manaar; (An. and Mag. Nat. Hist. 
 S. 5, Vol. 6, p. 478, PL V, fig. 1 8). Cylindrical spicules 
 have also been figured by Oscar Schmidt in a sponge from 
 the Atlantic, Suberites arciger (Atlant. Spong. p. 47, Taf. V, 
 fig. 6). Cylindrical spicules similar in form but of larger 
 size have lately been discovered by Mr. Carter in Irish Carboni- 
 ferous strata (An. and Mag. S. 5, Vol. 6, p. 212, PI. XIV, 
 fig. 14). They are present also in the Chalk Strata of Coes- 
 feld. (Ueber Coelop. p. 40, Taf. IV, figs. 38 43) and in the 
 Green Sand at Haldon (An. and Mag. Nat. Hist. S. 4, Vol. 7 
 PI. IX, fig. 53). 
 
 Scolioraphis? sp. 
 
 (Plate I, fig. 5). 
 
 Very elongated, sinuous spicules, apparently cylindrical, 
 of the same thickness throughout their length, irregularly bent 
 and fractured at their extremities. A central canal is exhibited 
 in some specimens. The surfaces of these spicules are far 
 more eroded than any others in this flint-meal, and forameni- 
 feral shells and other small objects frequently adhere to them, 
 so that they have an altogether different aspect to any other 
 
24 
 
 of the sponge remains, and at first led me to doubt whether 
 they were spicules, but the presence of the interior canal 
 points them out to be such. With one exception, in which 
 there is a slight tapering at one extremity, all these spicules 
 terminate abruptly, as if broken off at their ends, and therefore 
 may be regarded as incomplete. One of the longest of these 
 fragmentary spicules is 5,175 mm. long by 0,225 mm. wide, 
 others are only 0,135 mm. in width, but their outer surfaces 
 are so eroded that originally they may have been considerably 
 thicker. It is quite a matter of conjucture as to the sponge 
 of which these were the skeleton spicules, as, so far as I am 
 aware, no sponge with similar spicules has yet been described. 
 In Scolioraphis anastomans Zittel (Studien liber foss. Spong. 
 p. 95, Taf. XII, fig. i) there are also sinuous spicules, but 
 they are provided with knobbed surfaces, and are much smaller 
 than these from Horstead. 
 
 Order Tetractinellidae, Marshall. 
 
 The characteristic spicule of this division of sponges has 
 four arms or rays, one usually much longer than the others, 
 radiating from a centre. Each ray is provided with an in- 
 terior canal which unites with those of the other rays and, 
 where the rays are bifurcated, the canal is also bifurcated and 
 a branch extends into each division. There are two principal 
 forms in which these quadriaxial spicules may be divided. In 
 one there is a great development of a single ray, which be- 
 comes an elongated gradually tapering shaft, pointed at one 
 end, whilst from the other and usually the thickest portion of 
 the shaft, there spring three short arms or rays in very va- 
 rying directions ; in some spicules prolonged backwards , fork 
 like, nearly in a line with the shaft itself; in others, at varying 
 angles between the perpendicular and horizontal ; whilst in other 
 spicules the rays curve backwards , in the direction of the 
 point of the shaft, and become anchor-shaped. These three 
 
25 
 
 capitate rays vary, in different spicules, very much in their 
 form and dimensions, some are very small, cylindrical and 
 blunted, others, at the point of junction with the shaft, are as 
 stout as the shaft itself and become gradually pointed, whilst 
 in other spicules the head rays are largely developed, fre- 
 quently accompanied by a diminution of the shaft, which in 
 the extreme forms is reduced to a small blunted process. 
 Neither are the rays themselves uniform, very often one is 
 longer or shorter than the others of the same spicule. Another 
 feature of these capitate rays is that one or all of them may 
 bifurcate and the divisions, like the simple rays, are often of un- 
 equal length. That this bifurcation of the capitate rays is 
 only a further development of the growth of each ray and 
 not characteristic of a particular form of spicule, is evident 
 by the numerous examples in this flint meal , in which a 
 gradual transition may be seen from specimens which have 
 all the rays simple, to those in which one, two, and finally all 
 three of the rays are bifurcated. 
 
 For this form of spicule in which the shaft is very pro- 
 minent in relation to the head rays, as well as for those in 
 which the shaft has been reduced, but can yet be recognized 
 as distinct from the head rays, I propose to employ the term 
 trifid spicule, and where the rays are simple, to denominate 
 the spicule simple trifid^ and where bifurcate, compound 
 trifid.* 
 
 The other main form of quadriaxial spicule has the four 
 divisions or arms more or less like each other, so that it is 
 not easy to determine which of them corresponds to the shaft 
 of the trifid spicule. In a very few instances one of the arms 
 of these spicules has been extended beyond the central point 
 so that a five rayed spicule results. The arms of these four 
 rayed spicules sometimes radiate from the centre at an equal 
 angle from each other, but in other examples one arm is at 
 a more or less open angle and occasionally nearly at a right angle 
 to the other three arms. A few of these spicules also have 
 
26 
 
 a tendency to bifurcate. I propose to designate this form of 
 spicule <tsimple or compound quadrifid^ according as the arms 
 are simple or bifurcate. 
 
 In addition to these trifid and quadrifid spicules, all or 
 nearly all the sponges of this order have acerate spicules, 
 often of two dimensions in the same species, and many have 
 a surface layer of minute globate and globo-stellate spicules 
 or small discoidal bodies. 
 
 All the different forms of spicules which have been re- 
 cognized in existing sponges of this order ; the globate, globo- 
 stellate and discoid bodies forming the outer cortex; the trifid 
 spicules of the <?;/<? supporting the crust ; the acerate and 
 quadrifid spicules of the frcdy] and the anchoring spicules, 
 are represented in the Horstead flint meal, and if we may 
 judge by the relative number of the spicules, the Tetractinel- 
 lid sponges must have far exceded the contemporary Lithistid 
 and Hexactinellid sponges of the Chalk. In existing sponges 
 of this order the skeleton spicules are distinguished by being 
 much more robust than those of other types, and the same 
 characteristic is displayed by the Chalk spicules. 
 
 As regards the acerate spicules which I regard as be- 
 longing to this group, there is no peculiarity by which those 
 belonging to different genera can be distinguished , nor are 
 the minute spheres, stellates and discs of much service either 
 in this respect, but it is a question how far the different forms 
 of trifid and quadrifid spicules, which compose the zone and 
 body spicules of these sponges, may be representative of 
 different species. According to Mr. Carter's descriptions of 
 existing species of Geodia and Stelletta from the South coast 
 of Devon, (An and Mag. Nat. Hist. S. 4, Vol. 7, p. 9, 
 PL IV) from the Atlantic, (Vol. 18, p. 397, PL XVI); and 
 also from the gulf of Manaar, (S. 5, Vol. 6, p. 485, PL V, VI), 
 there are present in each species, in addition to the globate and 
 stellate spicules, the large and small acerates, and the minute 
 anchor-shaped trifid, a single form of large trifid which pre- 
 
27 
 
 sents characters sufficient to distinguish it from the large 
 trifids of other species. Dr. Bowerbank also figures a single 
 large trifid as one of the characteristic spicules of Geodia 
 Zetlandica (Monogr. Brit. Spong. Vol. Ill, p. 15, PL 7, fig. 6). 
 On the other hand O. Schmidt in his descriptions of species 
 of Geodia and Stelletta from the Adriatic (Spong. d. Adriat. 
 Meeres 1862, p. 46, 49, 50, Taf. Ill, IV) represents two or 
 three forms of large trifid spicules (Anker) in the same species 
 and some of these appear common to two or more different 
 species. It thus becomes very uncertain whether the numerous 
 and very variable forms of large trifid spicules which are pre- 
 sent in this flint-meal should be taken to indicate as many 
 species of sponge, or whether two or three of different forms 
 may have belonged to one and the same species. Even 
 however, if we accept this latter view, it will be seen from 
 the number of forms of these spicules , that a considerable 
 number of species of this group lived together in the Cretaceous 
 ocean. Before describing these trifid spicules in detail, I pro- 
 pose first to refer to the aceralte spicules which probably 
 accompanied them. 
 
 Acerate Spicules of Geodia and allied Genera. 
 
 (Plate I, figs, i 3, 20, 21). 
 
 Comparatively large, simple, straight and curved spicules, 
 which may be arranged under three different forms; First, 
 Straight spicules (fig. i) thickest in the centre and gradually 
 tapering towards the pointed extremities ; Second , Curved 
 spicules (fig. 2) also pointed at both ends, and thicker in pro- 
 portion to their length than the straight specimens ; Third, 
 Spicules of a rod-like character (fig, 5) nearly cylindrical or 
 diminishing very gradually towards the extremities ; I have 
 found none of these latter retaining their ends perfect. Sur- 
 face of these spicules apparently smooth originally. In none 
 has the central canal been preserved. The shorter forms of 
 these spicules measure 3,37 mm. long by 0,045 mm. wide, 
 
28 . 
 
 whilst the larger attain dimensions of 5,5 mm. in length by 
 0,135 mm. in width. Occasionally spicules of still larger di- 
 mensions are present; one of these, though incomplete at 
 one end, measured 9,2 mm. long by 0,214 mm. wide; a frag- 
 ment of another was 0,35 mm. in diameter. 
 
 There are also straight and curved spicules (figs, 8, 9) 
 facsimiles of the larger spicules as regards their form but 
 very much smaller. The straight forms of these small exam- 
 ples average 1,35 mm. long by 0,022 mm. wide, whilst the 
 curved, shorter and proportionally stouter, average 0,585 mm. 
 in length by 0,067 mm - m width. These curved spicules are 
 abruptly pointed at both ends and very constant in form and 
 size. Both the larger and smaller forms are abundant in the 
 flint meal. 
 
 Simple acerate spicules are common to so many different 
 forms of sponges that it is impossible to determine the rela- 
 tions of detached spicules with certainty, but, judging from 
 the size of those in this deposit, they appear to me to cor- 
 respond rather with acerates which in existing sponges are 
 associated with the trifid spicules of Geodia and Stelletta 
 than with the acerates of monactinellid sponges. Similar 
 spicules are described by Mr. Carter from the Haldon Green 
 Sand and referred to Geodia - like sponges (An. Mag. Nat. 
 Hist. S. 4, Vol. 7, p. 129, PI. 10, figs. 75, 76); they are 
 present in the Upper Chalk of Coesfeld (Ueber Coelop. p. 36, 
 Taf. IV, figs. 14) in the Chalk of the North of Ireland 
 (Wright: op cit. Plate II, fig. i) and also in the Eocene sand 
 of Brussels (Rutot, op. cit. PL III, figs. I 4). 
 
 Ophiraphidites, sp. 
 
 (Plate I, figs. 69). 
 
 Acerate spicules gradually tapering from the centre to 
 either end; sinuous and curved in a great variety of forms. 
 Some specimens have the main portion of the spicule nearly 
 straight and the extremities abruptly bent in opposite direc- 
 
29 
 
 tions, in other instances they are bow-shaped or curved in 
 the form of the letter C. One of the largest specimens is 
 3,8mm long by 0,13 mm wide; an average specimen measures 
 2,25 mm long by 0,09 mm. wide. They are very abundant. 
 
 These spicules correspond pretty closely with those which 
 Mr. Carter discovered on the outer surface of some Atlantic 
 sponges, to which he has provisionally given the name ol 
 Ophirapkidites tortuosus, it being at present doubtful whether 
 they may belong to a distinct sponge or whether they are the 
 acerates of a species of Pachastrella. (An. Mag. N. H. Dec. 
 1876 p. 458). Similar spicules are also present in the Haldon 
 Green Sand, (An. Mag. N. H. Vol. 7 p. 131 PI. X. fig. 79); 
 in the Westphalian Chalk, (Ueb. Coelop. Taf. IV. fig. 24, 25) 
 and in the Eocene sand at Brussels. (Rutot: op. cit. PI. 3, 
 figs. 5, 29). 
 
 Simple and Compound Trifid Spicules of Geodia and 
 
 allied Genera. 
 Geodia? clavata n. sp. 
 
 (Plate I. fig. 4. plate II. figs 15.) 
 
 Very robust spicules with an elongate, straight or slightly 
 curved shaft, constricted immediately beneath the head ; below 
 this, there is a bulbous inflation from which the shaft gradually 
 tapers to a point. The head rays in all the simple forms, and 
 in some of the compound as well, are but little more than 
 short rounded knobs (fig. I 35); in other compound heads 
 the rays are more compressed and bifurcate into small pointed 
 extensions. In some instances the number of the rays is 
 limited to two. (fig. i). The interior canal of the shaft is 
 partially preserved ; in some spicules it may be seen to extend 
 throughout the entire length; in others only traces of it are 
 exhibited in the inflated portion of the shaft, where it is 
 expanded in the same manner as the shaft itself. Very 
 rarely are canals exhibited in the head rays. In some spicules, 
 which at first I thought might be acuate spicules with bul- 
 
30 
 
 bous heads , but which I now regard as trifids in which the 
 head rays have been broken off (Plate I, fig. 4), the canal is 
 now extremely wide ; but in the majority of the specimens 
 the canal is partially or wholly infiltrated with silica. The 
 surface of these spicules is very rough, and the peculiar 
 radiate structure of the silica of the interior is well displayed. 
 Scarcely any of these spicules are complete, the lower part 
 of the shaft being mostly absent. A small specimen with 
 the shaft complete has a length of 7 mm.; others in which 
 the terminal portion is wanting, measure 8.77 mm. long, 
 with a width of the bulbous portion of the shaft of 
 0,585 mm. They are not uncommon. 
 
 The gradual development of the rays of the head is 
 well shown in different examples of these spicules. In one 
 instance there are but two of these knob shaped rays, as if 
 the summit of the shaft had been merely bifurcated (fig. i) ; 
 then follow specimens in which three simple knobs or rays 
 are displayed (fig. 2), other specimens have one or more of 
 these knobs but slightly constricted (fig. 3) whilst in others, 
 the bifurcation of the rays is carried still further , the primary 
 rays are somewhat compressed and flattened, and each extended 
 into two short, obtusely pointed arms (fig. 4). The numerous 
 transition forms would seem to indicate that these variously 
 headed spicules may all have belonged to the same species 
 of sponge. 
 
 I am unable to find any description of fossil or existing 
 sponges with similar spicules; their form would indicate that 
 they were probably the 'zone' spicules of a sponge like Geodia, 
 and as they are distinct both in size and in the peculiar knob- 
 like character ot the head rays, I propose to designate them 
 Geodia ? clavata. Spicules with a constricted neck, but with a 
 somewhat different form of the head rays, and besides of much 
 smaller dimensions are figured from the North of Ireland. 
 (Wright: op. cit. pi. II fig. 14). 
 
Geodia? coronata n. sp. 
 
 (Plate II fig. 6, 7, 8.) 
 
 Very robust spicules with a conical shaft, inflated at the 
 upper end, and from thence gradually tapering- to a point. 
 Immediately on the inflated shoulder of the shaft, or with 
 a very short intervening neck, are placed from 3 to 6 relatively 
 very short, pointed rays, at a more or less open angle with 
 the shaft. The shafts of these spicules are mostly straight, 
 but in one example the lower end is curved. The rays, both 
 of the simple and compound forms are very equal in size, 
 and the compound rays bifurcate so close to the head of 
 the shaft, that, in many instances, they appear to spring from it 
 independently. As in the previous species , one or all of 
 the primary rays of a spicule may become bifurcated. The 
 total length of a small spicule (fig. 8) is 2,25 mm.; width 
 across the head 0,45 mm.; diameter of shaft 0,27 mm. The 
 length of an average spicule is 4,05 mm., the width of the 
 rays 0,31 5mm.; and of the shaft 0,45 mm. These spicules 
 are somewhat rare. 
 
 From the peculiar arrangement of the rays, I propose 
 to name this spicule Geodia ? coronata. .Similar spicules are 
 present in the Chalk of the North of Ireland (Wright: op. cit. 
 PL II fig. 133, 
 
 Geodia? Wrightii n. sp. 
 
 (Plate II fig. 12). 
 
 Robust simple trifid spicules with an elongated shaft 
 gradually tapering from the summit to the pointed extremity, 
 the arms straight and directed forwards at angles varying 
 between 35 and 60 with the shaft. The shaft furnished 
 with from 10 to 14 ring-shaped expansions, extending at 
 regular intervals from each other, from the summit to two-thirds 
 or three-fourths of its length, the lower portion of the shaft being 
 smooth. Each of the arms has also 4 to 6 similar rings. In some of 
 the smaller spicules the rings of the shafts and arms are by no 
 
32 
 
 means so well defined as in the larger, and forms occur which 
 show only traces of the rings, but I am unable to determine, 
 whether this is owing to their non development or to subse- 
 quent erosion in these smaller specimens. The simple interior 
 canal is shown in the larger forms. Length of an average 
 large specimen 2mm, width of shaft 0,247 mm -'> width across 
 the head 0,45 mm. Not very common. 
 
 I have named the spicule in honour of Mr. Joseph Wright 
 F. G. S. who first discovered this form in the Irish Chalk 
 (List of the Cretaceous Microzoa, Plate II fig. 5). In the 
 figures of the Irish specimens, the rings extend quite to the 
 extremity of the shaft, and in this respect differ slightly from 
 the English forms. I have discovered similar spicules in material 
 from Coesfeld and they are also present in beds belonging to 
 the Lower Green Sand (Neocomian) in Surrey. Quadrifid spicules 
 and also curved cylindrical spicules with similar ring-shaped 
 expansions are described by Mr. Carter from the Haldon Green 
 Sand under the names of Monilites quadriradiatus and 
 M. Haldonensis respectively (An. Mag. N. H. S. 4 Vol. 7 p. 
 132 pi. IX, fig. 44 47) but the trifid form does not appear 
 to occur in these beds. An annulated quadrifid spicule has 
 also lately been discovered by Mr. Carter in an existing sponge, 
 Tisiphonia annulata (An. Mag N. H. S. 5 Vol. 6 p. 494 pi. V. 
 figs. 28 a. d.) 
 
 In addition to these trifid spicules, I have also found in 
 the Horstead flint, a simple annulated acuate spicule similar 
 in all respects to the trifid spicules but minus the capitate 
 rays. At first I thought it might have been perfect for similar 
 acuate forms are common in Chalk Flints in Oxfordshire, and 
 also in the North of Ireland, but on examination under the 
 microscope, I find that the head of the shaft is not evenly 
 rounded off and that therefore the spicule is merely a headless 
 trifid. 
 
33 
 Stelletta? sp. 
 
 (Plate II fig. 9, 10.) 
 
 Robust spicule with straight gradually tapering shaft and simple 
 trifid head; the rays stout, short and recurved. In one example 
 the shaft has a contracted neck ; in others, the shaft is thickest 
 immediately beneath the rays. The incomplete spicule (fig. 9) 
 has a length of 4,04 mm ; width of head rays 0,63 mm ; thickness 
 of shaft 0,382 mm. The smaller but complete spicule (fig. 10) 
 is 2,76mm long; width of shaft 0,157 mm. The interior 
 canals are partly preserved. These spicules are extremely rare. 
 
 Spicules similar in shape but much smaller are figured 
 by O. Schmidt in Stelletta Grubii (Adriat. Spong. Taf. IV 
 fig. 2 a, b) and Dr. Bowerbank represents the 'zone' spicule 
 of Stelletta (Ecionemia) coactura (Mon. Brit. Spong. Vol. Ill 
 p. 269 PI. 82 fig. 1 6) with similar short curved head rays. 
 This spicule occurs also in the Irish Chalk (Wright. : op. cit. 
 PL II figs. 8. 9) and is referred by Bowerbank to Tethea and 
 Geodia. 
 
 Stelletta? sp. 
 
 (Plate II figs. II, 1 1 a.) 
 
 Robust spicule with elongated gradually tapering shaft 
 and head of simple conical, pointed, relatively very short rays, 
 directed obliquely forwards. Length of specimen figured 
 3,98mm., thickness of shaft 0,24 mm. ; width across the rays 
 0,5 4 mm. Rare. 
 
 A spicule of similar form but only half the size is met 
 with in Stelletta Boylicii, O. Schmidt (Sp. d. Adriat. Meeres 
 p. 47 Taf. IV. fig. 4). The majority of the 'zone' spicules 
 of Geodia, Stelletta, etc., have the head rays much more 
 extended than is the case with this form of spicule. This 
 form has also been met with in the North of Ireland (Wright : 
 op. cit. Plate II fig. 9.) 
 
 3 
 
34 
 Geodia? sp. 
 
 (Plate III. figs. 3, 6, 7, 9;- 
 
 Moderately robust, simple and compound trifid spicules, 
 with elongate, straight or slightly curved shaft. The head rays 
 of the simple forms are short, conical, pointed and directed 
 obliquely forwards; in the compound spicules the rays are 
 stouter and more compressed, with the primary rays directed 
 forward, but the secondary bifurcations extend in a horizontal 
 direction. There are considerable differences in the dimensions 
 of these spicules and the figures represent specimens of the 
 average largest of each form. The smallest (fig. 9) has a 
 length of 1.28 mm.; thickness of shaft 0,135 mm.; and 
 width of head rays 0,405 mm. The larger, (fig. 3) has a 
 length of 3,15 mm.; width of shaft, 0,22 mm. and width of 
 head rays 0,742 mm. There are a number of examples of 
 each of the four forms which I have figured and the exam- 
 ples of each form correspond pretty closely in size, so that 
 it is quite possible that they may belong to more than one 
 species of sponge. Spicules similar to these occur in other 
 groups of sponges besides Geodia and genera allied to it. 
 Thus a form similar to that placed under fig. 3 is found in 
 OphirapJiidites cretaceus, Zittel (Stud. lib. foss. Spong. p. 98, 
 Taf. XI, fig, 2 d) and it is also present in great numbers in 
 the Lithistid sponge Dorydernia dichotoma, Roemer, sp. (Stud, 
 lib. foss. Sp. Taf. VII, fig. I c) but as none of the spicules of 
 the mesh of Doryderma are found in the flint-meal, these tri- 
 fids are not likely to have belonged to this species. Similar 
 spicules are very abundant in the Haldon Green Sand and 
 are included by Mr. Carter under the name of Geoditcs hal- 
 donensis (An. and Mag. Nat. Hist. Vol. VII, PL X, figs. 60, 
 61, 63, 66): they have also been figured by Zittel from the 
 Upper Chalk of Coesfeld. (Ueber Coelop. Taf. VI, fig. 6, 9, 
 Taf. VII, figs. 2, 3, 6). 
 
35- 
 Geodia? sp. 
 
 (Plate II, figs. 13, 14, 15). 
 
 Spicules with an elongated, moderately robust, straight, 
 shaft, having a constriction at the upper end, below which, 
 in some specimens, there is a slight inflation; the shaft 
 gradually tapers to a point. The rays of the head are short, 
 slightly curved at their junction with the shaft, then straight 
 and pointed. They extend forwards nearly in a line with the 
 shaft itself. Specimens with the rays bifurcate or compound 
 are equally as numerous as those with the rays simple. These 
 spicules are well characterized by the constriction at the top 
 of the shaft and by the manner in which the head-rays ex- 
 tend directly forwards. The length of an average specimen 
 is 2,29 mm.; thickness of shaft 0,135 mm.; and width across 
 the head 0,2 mm. A spicule with a similar diposition of the 
 head rays but apparently a larger form, is figured from the 
 North of Ireland (Wright: op. cit. PI. II, fig. 22). 
 
 This and a few other forms of spicule, described below, 
 in which the head rays are either directed forwards or else 
 recurved, appear to resemble those spicules of Mr. Carter's 
 group of the Pachytragidae which he has termed anchoring* 
 spicules and it is probable that they were associated with 
 the larger trifid spicules, which I regard as the zone spi- 
 cules of the same division of sponges. As a rule these fossil 
 fork and anchor spicules are much larger than those in 
 existing sponges, but this might be expected from the greater 
 dimensions of the zone spicules. 
 
 
 Geodia, sp. 
 
 (Plate II, figs. 17, 18, 19). 
 
 Somewhat delicate spicules with an elongate gradually 
 tapering shaft, and simple, straight or curved, pointed rays 
 extending forwards from the shaft. The rays are frequently 
 
unequal in length und occasionally one is aborted and the spi- 
 cule is bifid. I have not noticed any specimens of this form 
 with compound or bifurcate rays. There is considerable va- 
 riation in the size of these spicules ; a large specimen (fig. 1 7) 
 measured 3,15 mm. long by 0,09 mm. wide. A smaller form 
 (fig. 19) is 1,01 mm. long and the thickness of the shaft is 
 0,045 mm - 
 
 A spicule closely resembling fig. 1 7 is figured by Dr. Bow- 
 erbank from Geodia Zetlandica (Mon. Brit, spong. Vol. Ill, 
 p. 15, PL VII, fig. 8). 
 
 Geodia? sp. 
 
 (Plate II, fig. 1 6). 
 
 Spicule with a delicate, straight, gradually tapering shaft 
 and three slender simple rays, unequal in length and recurved, 
 each of the rays being of nearly the same thickness as the 
 shaft, Length (incomplete) 1,05 mm. ; width 0,067 mm. Very 
 rare. This spicule is of the same character as the preceding, 
 but the arms are recurved, anchor-like. 
 
 Tethya? sp. 
 
 (Plate III, figs. I, 2, 13, 14). 
 
 Simple trifid spicules with a relatively very variable length 
 of shaft, which is straight or curved, gradually tapering and 
 in most instances obtusely pointed; the arms of the head 
 rays are very long, frequently of unequal length in the same 
 spicule; stout, conical, curved, pointed and directed forwards 
 at a very open angle with the shaft. There is a very great 
 amount of variation in the length and in the relative propor- 
 tion of the arms of these spicules; in some examples, one of 
 the rays becomes nearly as long as the shaft. In no instance 
 do the arms bifurcate. The length of one of the largest 
 specimens (fig. 2) is 4,025 mm.; width of shaft 0,247 mm,; 
 width of head rays 1,35 mm. In another form (fig. 13), in 
 which the shaft is constantly straight and but little stouter 
 than the arms, the total length of the spicule is 1,8 mm. dia- 
 
37 
 
 meter of the shaft 0,18 mm.; and the width across the head 
 rays 1,37 mm. The length and spreading disposition of the 
 rays give these spicules an altogether different appearance to 
 those which I have already referred to Geodia. They are not 
 uncommon in the flint-meal. 
 
 The only fossil spicules, which, to my knowledge, at all 
 resemble these specimens, are those of Tethyopsis Steinmanni, 
 Zittel (Stud, iiber foss. Spong. p. 99, Taf. XI, fig, 3) from the 
 Upper Chalk at Ahlten in Hanover. 
 
 Tethya? sp. 
 
 (Plate III, fig. 10). 
 
 Simple trifid spicules with elongated gradually tapering 
 shaft and three equal, short, conical, slightly curved and pointed 
 rays, projecting forwards at a wide angle with the shaft. Average 
 length of large spicule 2,47 mm.; diameter of shaft 0,135 mm.; 
 width across the head rays 0,83 mm. Not uncommon. 
 
 Tethya? sp. 
 
 (Plate III, fig. 4). 
 
 Simple trifid spicule, shaft straight, stout, rapidly taper- 
 ing; head rays short, cylindrical and nearly at right angles 
 to the shaft. Length of spicule 2,29 mm.; diameter of shaft, 
 0,315 mm. extension of head rays 0,8 mm. This peculiar 
 shaped, robust spicule occurs but rarely. It has been found 
 in the Chalk of the North of Ireland (Wright: op. cit. p. 89 
 PL II, fig. 12) and Dr. Bowerbank states that a similar form 
 is present abnormally in some species of Tethya, but at pre- 
 sent I have failed to find any representation of a similar 
 spicule. 
 
 Tethya? sp. 
 
 (Plate III, fig. 5).' 
 
 Robust simple trifid spicule with stout, straight, tapering 
 shaft and conical, straight, moderately long, obtusely pointed 
 
- 38 - 
 
 arms, which are extended nearly at right angles to the shaft. 
 Length 2,07 mm. ; thickness of shaft 0,2 mm. and extension 
 of the head rays 1,08 mm. Very rare. 
 
 Tethya? sp. 
 
 (Plate III , fig. 8). 
 
 Simple robust trifid spicule, with a very stout, short, ra- 
 pidly tapering shaft, on the summit of which are three short 
 cylindrical rays, directed forwards at an acute angle with the 
 shaft. Length of shaft 1,95 mm.; thickness 0,27 mm.; ex- 
 tension of head rays 0,517 mm. This spicule is distinguished 
 from the preceding by the arms being longer, not so stout 
 and also directed forwards. Rather rare. 
 
 These three last described spicules (Plate III, figs. 4, 5, 8) re- 
 semble each other in having comparatively short and robust shafts 
 and cylindrical, stout, simple, head rays, at the same time each 
 form is distinctly marked off from the others by differences 
 in the shaft and disposition of the head rays, so that they 
 may have belonged to as many species of sponge. This form 
 of spicule differs considerably from the zone spicules of 
 existing sponges of the genera Geodia, Stelletta and Tethya, 
 nevertheless their affinities appear to approach nearer to the 
 spicules of this, than of any other group of sponges. 
 
 Siliceous Globules and Globo-stellates of Geodia 
 and allied Genera. 
 
 (Plate I , figs. 25 , 26 . 27). 
 
 There are present in the contents of this Horstead flint, 
 three different forms of the small siliceous bodies which are 
 so abundant in the dermal crust of the recent sponges belonging 
 to Geodia and its allied genera. The first of these (fig. 25) 
 are small bodies, either nearly spherical or ellipsoidal in form, 
 and showing, in many examples, an irregularly shaped cavity 
 
39 
 
 in the interior. The surface is now rough but appears to have 
 been originally smooth. An average specimen of the ellip- 
 soids, which is the prevailing form, measures 0,292 mm. and 
 0,247 mm. in the longer and shorter diameters respectively. 
 I have not noticed any trace of depression or hilum in any 
 of these bodies, which are abundant. 
 
 There can scarcely be any doubt that these small ellip 
 soids belong to the dermal crust of a sponge similar to Geodia. 
 The beautiful radiate arrangement of the siliceous balls of 
 the recent sponges has however, through the alteration of 
 the silica, quite disappeared in these examples, but I have 
 had the opportunity of comparing them with similarly shaped 
 spicules from the Upper Chalk of Coesfeld in which the ra- 
 diate structure is still preserved equally as perfect as in those 
 of existing sponges. In the Coesfeld examples I have noticed 
 a small cavity in the cental portion of the spicule, into 
 which the radiate structure does not extend, and to the en- 
 largement of this originally small cavity may be attributed 
 the irregularly shaped interior hollows of the Horstead spe- 
 cimens. This is further confirmed by finding in one of the 
 Coesfeld examples, an abnormally enlarged central portion, 
 while the exterior portion of the spicule yet retains the ra- 
 diate structure. 
 
 Excellent illustrations of the radiate structure of these 
 bodies in the existing sponge Geodia Me Andrcwii , Bower- 
 bank are given by Dr. Bowerbank (Mon. Brit. Spong. Vol. I, 
 PL XXIII, XXIV, figs. 326 335. Similar siliceous globules 
 are described by Mr. Carter from the Haldon Green Sand 
 (An. Mag. Nat. Hist. Vol. VII, PL IX, figs. 55. 56); by Prof. 
 Zittel from Coesfeld (Ueber Coelop. Taf. V, figs. 1820); 
 and also by Rutot from the Eocene Sand of Brussels (op. 
 cit. PI. Ill, figs. 36, 37). 
 
 The second form of these siliceous bodies (fig. 26), which 
 Mr. Carter has termed globo-stellate consists of a minute 
 spherical body whose surface is thickly studded with very 
 
40 
 
 minute pointed spines. They average 0,18 mm. in diameter. 
 In existing sponges a somewhat similar form of globo-stellate 
 is present in Stelletta lactca, Carter. (An. Mag. Nat. Hist. 
 S. 4, Vol. VII, p. 9, PL IV, fig. 21 a) and also in Tethya 
 robusta, Bowerbank. (Mon. Brit. Spong. Vol. I, PL VI, 
 fig. 165) and it is probable that these Horstead examples be 
 long to one or other of these genera. O. Schmidt has 
 also noticed the similarity of the globo-stellates of Tethya and 
 Stelletta. (Spong. d. Kiiste von Algier. p. 21). Globo-stel- 
 lates precisely similar to those from Horstead are also pre- 
 sent at Coesfeld. (Ueber Coelop. Taf. IV. figs. 28, 30). 
 
 In the remaining form of globo-stellate (fig. 27) the spi- 
 cule is larger and the rays are proportionately longer and 
 not so thickly set on the surface as in the form just described. In 
 an average specimen, the diameter, including the rays, is 0,29 mm. 
 and the rays have a length of 0,0675 mm. Varied forms of 
 these globo-stellates appear to be present in the same specimen 
 of sponge and these spicules probably belong to the same 
 species as the preceding. A similar body is figured by Rutot 
 from the Eocene of Brussels, (op. cit. PL III, fig. 35). 
 
 Free Siliceous Discs of Stelletta? 
 
 (Plate I, figs. 23, 24, 28). 
 
 Besides the minute globates and globo-stellates, there are 
 present in the Horstead flint, two forms of small, flattened 
 disc like spicules. One of these (fig. 23) is an extremely 
 delicate, thin disc, circular in outline, with a well defined 
 smooth border and an apparently smooth outer surface. 
 Arranged at regular distances round the border of the disc 
 is a series of about 18 extremely delicate flask-shaped canals, 
 having their narrow end directed to the edge of the disc, which, 
 however, they do not reach, and extending somewhat more 
 than one third of the distance to the centre. Between each 
 of these canals, there are two, and occasionally three, still 
 
smaller oval canals arranged at the same distance from 
 the edge of the disc. There is no trace of any canals radiating 
 from the centre. These discs are tolerably uniform in size 
 and have an average diameter of 0,315 mm. 
 
 The other form of disc (fig. 24) is likewise extremely 
 thin; elliptical in form, and having the outer edge deeply 
 notched or scalloped at regular intervals. In most of the 
 specimens the canals are obliterated , but one fragmentary disc 
 (fig. 26) shows about 18 straight canals radiating from the 
 centre towards the circumference. The long diameter of these 
 specimens measures 0,27 mm., and the shorter 0,1 8 mm. Both 
 forms are rare. 
 
 Professor Zittel has described and figured similar discs 
 from the Upper Chalk of Westphalia , (Ueber Coelop. p. 47. 
 Taf. V. figs. 32 35) and he is disposed to regard the discs 
 with the scalloped edges as the incomplete forms of those 
 with the even border. The Westphalian specimens lend 
 greater probability to this supposition than the Horstead 
 examples, for some of the even bordered discs from West- 
 phalia have, in addition to the flat and oval shaped canals 
 near the border of the disc, a variable number of canals 
 radiating from the centre to the circumference similar to those 
 in the scalloped discs. In the Horstead specimens, on the 
 other hand, the canals radiating from the centre only appear 
 in discs of this latter form, and as these are all elliptical in 
 outline as well, in contrast to the circular even bordered 
 discs, I think that these two forms of discs may be considered 
 as separate from, though closely allied to each other. 
 
 Similar discs with numerous radiating canals have not 
 yet been discovered in existing sponges, so that there is no 
 direct clue to the affinities of these small bodies. Mr. Carter 
 discovered two of the even bordered discs in the Haldon 
 Green Sand and regarding them as allied to the circular and 
 branched discs which form the surface spicules of Disco- 
 dermia and other allied Lithistid sponges, named them 
 
42 
 
 % 
 
 Dactylocaly cites callodiscus (An. Mag. N. H. S. 4 Vol. 7 p. 
 123 PL IX. fig. 40). It seems to me however, that the 
 affinities of these discs is rather with the siliceous globules 
 and stellates of Geodia and other genera of Tetractinellid 
 sponges than with the Lithistidae. In the surface spicules of 
 both recent and fossil Lithistids, in which canals are present, 
 they are restricted to three, which afterwards bifurcate in some 
 instances. In no known instance of ascertained Lithistid sur- 
 face spicule is there the number and disposition of the canals 
 as in these isolated discs. Considerable light is thrown upon 
 the affinities of these discs from the description which 
 O. Schmidt has given of the different stages of growth of 
 certain thin oval discs present in the existing sponge Stelletta 
 Euastrum (Spong. d. K. von Algier, p. 20, Taf. IV, fig. 4). 
 In the earliest stage of the development of these discs there 
 is a granular centre from which a single layer of extremely 
 delicate needle-like rays or spicules radiate. More developed 
 specimens show these fine ray-like spicules united near their 
 bases and deeply notched at the periphery. In the complete 
 disc one surface is smooth, flat, or slightly concave, whilst 
 the other is convex and thickly covered with small warts. 
 
 The structure of these discs from the Upper Chalk may 
 be understood, if we suppose that in a similar manner to 
 those of Stelletta Euastrum, they have been developed by 
 the amalgamation of a single layer of very fine spicules ra- 
 diating from a centre and that the shorter and longer canals 
 which are shown in the discs are those of the spicular rays 
 of which they are composed. Although O. Schmidt does not 
 mention the presence of canals in the spicular rays of the 
 discs of Stelletta Euastrum, he yet states that he has repeat- 
 edly noticed a central canal in the similar and equally mi- 
 nute spicular rays of the siliceous globules of Geodia canali- 
 culata (Spong. d. K. von Algier, p. 21). There appears there- 
 fore a probability that these small bodies are the surface discs 
 of a species of Stelletta. 
 
43 
 / 
 
 Tisiphonia? sp. 
 
 (Plate III, figs. 1623). 
 
 Compound trifid spicules, with stout, straight, more or 
 less developed shaft and very robust head rays, which are 
 bifurcated and widely extended in a horizontal direction. As 
 will be seen by the figures, there is great variation in the 
 relative dimensions of the arms and the shaft in these spicules. 
 In some specimens (fig. 17) the shaft is comparatively long 
 and the head rays stout, and but moderately expanded. The 
 example figured has a length of 2,115 mm -> tne sna ft is 
 0,112 mm. in diameter, and the extension across the head 
 1,26 mm. Another spicule (fig. 18) has a shorter and thicker 
 shaft and also stouter head rays. Figs. 16 and 20 show 
 forms in which the head rays are slender aud more divergent. 
 In fig. 21 the shaft is truncated and rounded at the end. In 
 fig. 22 the shaft resembles a stout cone and the head rays 
 are also very robust and obtusely pointed. This specimen 
 has a length of 1,35 mm.; thickness of shaft 0,247 mm. and 
 the extension of the head 0,967 mm. Lastly in figs. 19 and 
 20 we have examples in which the head is largely developed 
 and the shaft ot the spicule has become reduced to little 
 more than a mere rounded prominence. The width of the 
 head in these specimens is 1,6 mm. The figures show also, 
 how unequally the head rays of these spicules are developed ; 
 in some examples the primary rays are symmetrical, compa- 
 ratively long and extend some distance from the centre be- 
 fore they bifurcate and then give off long pointed secondary 
 rays (figs. 16, 20); in other specimens the primary rays 
 are so wide that they amalgamate and form a solid laminated 
 centre from which the unequally developed, obtusely blunted 
 secondary rays are given off. In many specimens the canal 
 system is very distinctly shown, those of the arms radiating 
 from the centre of the shaft and sending a branch to the 
 extremity of each ray. 
 
44 
 
 X 
 
 Such differences of size and form indicate that these spi- 
 cules may have belonged to several species of sponge and 
 this is rendered further probable by the great numbers which 
 are present in this Horstead flint. 
 
 I have some difficulty in determining the relationship of 
 this group of spicules with existing forms. The compound 
 trifid spicule with the head expanded horizontally frequently 
 forms the zone spicule of Geodia and Stelletta, but accom- 
 panied by an elongated shaft; a similar form of spicule is 
 present in the dermal portion of many species of Lithistids, 
 but these are so diminutive in comparison with those figured 
 as to negative the probability of their being surface spicules of 
 these sponges ; I also compared them with the truncated spicules of 
 Pachastrella, but these have the arms mostly simple, or where bifur- 
 cated, the division is generally restricted to the ends of the arms ; at 
 last I found the description of a sponge with a similar spicule, 
 placed by Mr. Carter under the name of Tisiphonia nana(An, Mag. 
 Nat. Hist. S. 5, Vol. 6, p. 492, PI. VII, fig. 43 a, b). This spicule 
 has a truncated shaft and an expanded head closely resem- 
 bling some of the Horstead forms. Further, O. Schmidt has 
 figured the head of a spicule similar to that of fig. 20 in a 
 sponge from the Adriatic under the name of Stelletta Hilleri 
 (Supp. d. Spong. Adriat. Meeres, p. 32, Taf. Ill, fig. 8) and 
 as according to Mr. Carter the genus Stelletta is equivalent 
 to Tisiphonia, there is a probability that these spicules with 
 truncated shafts may be the zone spicules ot sponges allied 
 to Stelletta, notwithstanding the great differences which they 
 present, in the comparative reduction of the shaft and the 
 extended development ot the head, from the typical zone 
 spicules of Stelletta itself. 
 
 Similar spicules but as a rule very much smaller in di- 
 mensions, have been found by Mr. Carter in the Haldon 
 Green Sand (An. Mag. Nat. Hist, S. 4, Vol. 7, PL IX, 
 figs. 30 36) ; they are also present in the Chalk of the 
 North of Ireland (Wright op. cit. PI. II, figs. 17 a, b) and 
 
45 . 
 
 Westphalia, (Zittel : iiber Coelop. Taf. VI, figs. 16 29) 
 and in the Eocene sand of Brussels (Rutot: op. cit. PI. Ill, 
 
 fig- 9)- 
 
 Quadrifid Spicules of Pachastrella. 
 Pachastrella, sp. 
 
 (Plate III, figs. 24, 25). 
 
 Simple quadrifid spicules in which three elongated arms 
 radiate at equal angles from a centre and form the outline 
 of a more or less elevated three-sided pyramid, from the apex 
 of which another arm generally shorter than the other three, 
 extends upwards. There are two very distinct forms of these 
 simple quadrifid spicules. In the larger (fig. 24) the three 
 arms form a very low pyramid ; in some examples they are 
 nearly in the same plane; -and the fourth arm, much shorter 
 than the other three, is nearly at right angles to them. The 
 other three arms are nearly equal, straight and pointed, and 
 in no instance which I have noticed are they bifurcated at 
 the extremities. Canals are shown in some examples, much 
 wider than those in similar spicules of recent sponges. These 
 spicules are abundant and very uniform in size. The average 
 length of the arm of a fair sized spicule is 1,035 mm - an d 
 its thickness near the centre is 0,09 mm. The vertical arm 
 has a length of 0,2 mm. 
 
 In the smaller form of quadrifid spicule (fig. 25) the four 
 arms are nearly equal in length and they are disposed so as 
 to form a more elevated pyramid. The average length ot 
 each arm is 0,45 mm. and the thickness near the centre 
 0,045 mm - These spicules are also very abundant. 
 
 From comparisons which I have made with specimens of 
 PacJiastrclla amygdaloides , Carter , and P. geodioidcs , Carter 
 (An. Mag. Nat. Hist., S. 4, Vol. 18, p. 406, PL XIV, figs. 22, 
 23) from the Atlantic, I find that there is a striking resem- 
 
- 46 - 
 
 blance in the spicules of the recent sponges and those from 
 the Horstead flint. Beyond the differences in the state of 
 preservation there is scarcely anything to distinguish the fossil 
 Chalk spicules from those of the existing sponge, P. amyg- 
 daloides which was dredged from a depth of 292 fathoms near 
 Cape S'- Vincent. 
 
 Similar spicules, as regards form, have been found in 
 the Irish Chalk (Wright: op. cit. PI. II, fig. 7 a, b) in West- 
 phalia (Ueber Coelop. Taf. V, figs. 54,' 55) and also in the 
 Eocene Sand of Brussels (Rutot : op. cit. PI. Ill, fig. 8). 
 
 Pachastrella Carter!, n. sp. 
 
 (Plate III, figs. 29, 30, 31). 
 
 Simple quadrifid spicules, with short robust , cylindrical 
 or conical, and obtusely rounded arms, forming a depressed 
 pyramid with a short arm at the apex. In figs. 29 and 3 1 , the 
 stout conical blunted arms vary from 0,47 mm. to 0,67 mm. in 
 length, and are 0,225 mm. in thickness. In the more regular spicule 
 (fig. 30) the arms are cylindrical and measure 0,337 mm - by 
 o, 1 1 2 mm. These spicules appear intermediate between those forms 
 of skeleton spicule with long pointed simple arms of which P. amyg- 
 daloides, Carter is the type, and those in which one or more of 
 the arms are bifurcated at the extremities as in the recent 
 sponge P. abyssi, O. Schmidt , and the fossil , P. primoeva, 
 Zittel. It is true that in both the recent and fossil Pacha- 
 strellid sponges, various forms and sizes of quadrifid spicules 
 are present in the same sponge, but as I have not noticed 
 any sponges with spicules having the short, blunted arms of 
 these examples, I think they may belong to a distinct species, 
 which I propose to name in honour of Mr. H. J. Carter F. R. S. 
 
 Pachastrella primaeva ? 
 
 Zittel: Stud, iiber Foss. Spongien p. 100, Taf. XII, fig. 4. 
 
 (Plate III, figs. 28, 32, 33, 34). 
 Spicules with four unequal arms, straight or slightly 
 
47 . 
 
 curved, of which three are disposed nearly in a plane or 
 forming a very depressed pyramid. One arm ot these three 
 is frequently longer than the others, slightly curved, and at 
 the extremity bifurcates into two small irregular rays (fig. 32, 
 33). The arm at the apex of the pyramid is short and obtusely 
 blunted and in some instances is continued beyond the apex 
 of the pyramid, so that a five-armed spicule is the result, 
 (figs. 28, 34), As in other respects these five-armed spicules 
 correspond closely with the four-armed, I am disposed to consider 
 them as mere variations and not as indicating a different species. 
 The length of the arms of these spicules varies between 0,495 mm. 
 and 1,125 mm - with an average thickness of 0,14 mm. The 
 examples of these quadrifid spicules are but rarely met with. 
 
 These spicules correspond so closely with those of 
 Pachastrella primoeva, Zittel, from the Upper Chalk of Ahlten 
 in Hanover, that, so far as can be judged from the zone 
 spicules , they may have belonged to the same species. In 
 the Ahlten sponge, the larger spicules are also accompanied 
 by great numbers of the smaller simple quadrifid spicules 
 similar to those of my fig. 25, and these smaller forms both 
 in the fossil and recent sponges seem common to different 
 species of Pachastrella. I have also seen a specimen of Pachastrella 
 from the Flamboro' Chalk, given by Mr. Carter to Professor 
 Zittel, in which the furcation of the skeleton spicule is carried 
 to a greater extent than in the Ahlten and Horstead examples, 
 and the spicule is also less robust. In the furcation of the 
 extremities of the spicule, the Chalk examples correspond 
 closely with the skeleton spicules of Pachastrella abyssi, 
 O. Schmidt, and also with P. intertexta, Carter. (An. Mag. 
 Nat. H. S. 4. Vol. 18. p. 409 PI. XV. fig. 41) dredged from 
 the Atlantic near Cape St. Vincent at a depth of 374 
 fathoms. Examples of the five-armed spicules also occur in 
 P. inter tcxta and in P. parasitica, Carter (An. Mag. Nat. Hist. 
 S. 4. Vol. 1 8. p. 410 PL XVI, fig. 50.) 
 
- 48 - 
 Pachastrella ? sp. 
 
 (Plate III, fig. 27.) 
 
 Spicules with three straight slender arms radiating at 
 equal angles and in one plane. The extremities of the arms 
 are not preserved, but they are nearly uniform in thickness 
 throughout, and frequently one is shorter than the other two, 
 but whether this is merely accidental I am unable to say. 
 Average length of arms o,6imm. and width 0,045 mm. Not 
 uncommon. 
 
 It is doubtful whether these three - armed spicules 
 are merely spicules in which a fourth arm has not been 
 developed or whether they belong to a different group of 
 sponges than the Pachastrellidae. Mr. Carter states that in 
 P. amygdaloides large triradiates are present in which the 
 fourth arm or shaft is only represented by a short extension 
 of the central canal inside the spicule; and P. conncctens, 
 O. Schmidt (Atlant. Spong. p. 65) is also stated to have large 
 three-rayed spicules; and similar forms are also represented 
 in Sphinctrella horrida , O. Schmidt (op. cit. p. 65) so that 
 it is not improbable that they may belong to a species or 
 Pachastrella. 
 
 Caminus? sp. 
 
 (Plate III, fig. 26.) 
 
 Spicule with three unequal arms in the same plane. 
 One arm is stout, short, and rounded at the extremity, the 
 others are elongated and gradually tapering. Length of shorter 
 arm 0,38 mm. ; thickness 0.067 mm - I have only met with 
 a single specimen. The only instance which I have seen of a 
 spicule of corresponding form is represented in Caminus apiarinni, 
 O. Schmidt, (Atlant. Spong. p. 71. Taf. VI. fig. 14) from the 
 coast of Florida. The scale to which O. Schmidt's figure is 
 drawn is not stated, so that I am unable to make a comparison 
 as to their relative dimensions. 
 
49 
 
 Order Lithistida, Oscar Schmidt. 
 
 In this order are comprised sponges whose skeleton is 
 built up of spicules which are united together by the inter- 
 locking of the extremities of their arms to form a more 
 or less open mesh or net work. In addition to the spicules 
 composing the mesh-work of the body of the sponge, there 
 are, in most, if not all the Lithistids, other spicules, frequently 
 of quite a different form to those of the mesh, which are 
 arranged on the outer surface of the sponge and may be 
 termed surface spicules in contradistinction to the mesh or 
 body spicules. In existing sponges of this order minute 'flesh' 
 spicules are also present, but these have not been preserved in 
 the fossil examples. 
 
 The intimate manner in which the spicules of the mesh 
 are interwoven together by the curiously modified extremities 
 of their arms , has enabled the sponges of this order to retain 
 in many instances their form complete in a fossil condition. 
 As remarked by Zittel (Studien p. 79) the spicules do not 
 usually fall asunder and become detached after the death of 
 the animal , and Mr. Carter has also recently stated that the 
 mesh spicules of existing Lithistids are, for the most part, so 
 locked together that even boiling in nitric acid does not cause 
 them to detach and fall asunder. (An. Mag. N. Hist. S. 5, 
 Vol. 6 p. 496). It is to this fact that the larger portion of 
 known fossil sponges belong to this order , though , if one 
 may judge from the proportion of the detached spicules of 
 Tetractinellid sponges in this Horstead flint, as well as at 
 Haldon and in Westphalia, the Tetractinellids may have been 
 equally as numerous in the same strata with the Lithistids, but 
 that owing to their skeleton being built up of disconnected 
 spicules, and the inevitable dispersion of these in nearly all 
 cases after the death of the animal, their remains have, up 
 to the present, been but little noticed. At the same time 
 that the skeleton of fossil Lithistid sponges has been very 
 
 4 
 
50 
 
 often preserved entire, there seems reason to believe that in 
 the majority of instances the skeleton of the sponges of this 
 order after the death of the animal, fell apart and became 
 resolved into its constituent spicules. Such at all events 
 appears to have been the case in the Horstead Chalk. In this 
 Chalk, so far as I am aware, not a single entire Lithistid sponge 
 has been discovered, and yet in the contents of this single 
 flint from the same locality there are inclosed such a number 
 and variety of detached Lithistid spicules as to make it evident, 
 that the order was represented by numerous species. Nor has this 
 disintegration of the Lithistid sponges in the Chalk been 
 brought about by mechanical influences, for the individual spicules 
 are entire and present only those alterations in their structures 
 which are common to all the spicules of this flint meal. Not- 
 with standing the great abundance of single spicules, there 
 are but few specimens in which two or more of these spicu- 
 les yet remain together in their natural connection. 
 
 In this order there is an extraordinary variety of form 
 in the individual spicules of the skeletal mesh-work, as also in 
 the manner in which they are combined together. In some 
 spicules, the plan of growth is similar to that of the four-armed 
 spicules of Pachastrella , in others , the form is so irregular 
 that no definite system of growth can be perceived. While, 
 to some extent, the difficulties attending the determination of 
 the affinities of the detached spicules of the Tetractinellid 
 sponges are also present in the case of the Lithistids, 
 yet the very peculiar form of the Lithistid spicules 
 and the varied methods by which they are attached to each 
 other to form the mesh of the skeleton, furnish reliable in- 
 dications of their relationships ; besides which the investiga- 
 tions of Zittel, Carter, O. Schmidt and Sollas have supplied 
 the means of comparison with a number of genera and species, 
 and thus in many instances enabled the systematic position 
 of the detached spicules to be satisfactorily ascertained. In 
 
some instances several different genera have the mesh spicule 
 so nearly alike that it is impossible to determine to which 
 of the group the detached spicules may belong, in other cases 
 the spicules belonging to a single genus are very clearly dif- 
 ferentiated. Professor Zittel has divided the order of Lithisti- 
 dae into four Families according to the structure of the spi- 
 cules of the mesh and two of these families are present in 
 the Horstead flint. 
 
 Family Megamorina, Zittel. 
 
 The Skeleton spicules of this group are comparatively 
 large, elongated, oftentimes branched bodies, mostly very irreg- 
 ular in their form. They are united together to form the 
 skeletal mesh either by having the concave expansions of the 
 terminal arms of the spicules, closely fitted against the arms 
 of adjoining spicules as in the Genus Lyidium , O. Schmidt ; 
 or by the intertwining of the extremities of the spicules round 
 each other as in the case of the genera Carterella, Zittel 
 and Isoraphinia, Zittel. 
 
 Genus Lyidium, O. Schmidt 1870. 
 Lyidium Zitteli n. sp. 
 
 (Plate IV, figs. 1-9). 
 
 Robust elongated spicules, simple or branched, of various 
 forms, with the extremities of the arms transversely expanded 
 and concave, so as to be attached to the surfaces of adjoin- 
 ing spicules. Average diameter of the arms of the spicules 
 o, 1 1 2 mm. 
 
 There is such an extraordinary diversity in the forms of 
 the spicules of this species, that in some hundreds of examples 
 which I have seen, I do not think two similar ones could be 
 found. The simplest form is a curved spicule with the two 
 ends transversely expanded (fig. 2) ; from this, the next stage 
 is a curved spicule with a projecting arm in the centre of 
 
52 
 
 the curve (fig. 4), then one or both ends of the simple spicule 
 are bifurcated ; another spicule has three arms radiating from 
 a common centre as shown in the central spicule ot fig. 3 ; 
 other spicules have a straight or curved main stem from 
 which one to four branches are given off at various angles 
 and in different directions, each branch, as well as the stem 
 itself having expanded terminations (figs. 5 , 7 , 9) , in other 
 spicules one end of the main stem is not expanded but ter- 
 minates in an obtuse point (figs. 6, 8); in short, there is an 
 endless multiplicity of form of which only the more striking 
 varieties are indicated in the figures. At the terminations of 
 the stems and branches of these spicules, there is a more or 
 less elongated expansion at right angles to the direction of 
 the branch, and this is hollowed out lengthways in such a 
 manner that the depression exactly fits the convex surface 
 of the arms of other spicules and tightly grasps them so as 
 to form a strong complicated network with irregular openings. 
 A few specimens occur yet showing the manner in which the 
 spicules of the mesh were- united together (figs. I, 3) but 
 these examples are very rare in comparison with the number 
 of isolated spicules. Nowithstanding the great variety of form, 
 the spicules are very uniform in thickness , and one may com- 
 pare the skeleton to a piece of complicated wire netting made 
 up of innumerable fragments , cut and twisted into different 
 forms but all from the same strand of wire and fastened to- 
 gether in a similar manner. The surface of the spicules ap- 
 pears to have been originally smooth, though now rough and 
 pitted, and the delicate edges of the expanded extremities have 
 in most instances lost their evenness of outline. It is per- 
 haps owing to the erosion of these extremities that the spi- 
 cules lost their hold and became detached from each other. 
 In one solitary instance only have I noticed a canal in the 
 interior of a spicule. 
 
 The type, and up to the present, the only described spe- 
 cies of the Genus Lyidium is a fragment of sponge dredged 
 
53 -- 
 
 up from the Atlantic, off the Coast of Cuba, at a depth of 
 270 fathoms, on which O. Schmidt, on account of the highly 
 characteristic form and arrangement of the spicules, constituted 
 the genus. I have had the opportunity of examining a frag- 
 ment and also mounted slides of the original specimen, which 
 were presented by O. Schmidt to Prof. Zittel, and have thus 
 been able to compare the fossil Chalk spicules with those of 
 the existing sponge. The spicules of the existing sponge 
 have the same diversity of form and are attached together by 
 the close fitting expanded extremities in a similar manner 
 to the Horstead flint spicules, but the mesh is of a more open 
 character and the thickness of the spicules is considerably 
 less, in a word, to continue the simile, the strand of wire is, 
 in this case, of a different thickness having an average diameter 
 of 0,045 mm - but little more than one-third the thickness of Ly- 
 idium Zitteli. It is on this difference of thickness of the spi- 
 cules and on the closer character of the openings of the mesh 
 that I venture to constitute these fragments and spicules into 
 a distinct species, which it gives me great pleasure to name 
 in honour of Professor Dr. Karl Zittel. 
 
 Though first noticed as an existent sponge, the genus 
 Lyidium appears to have had a wide distribution in the past. 
 Mr. Carter has figured a number of spicules from the Haldon 
 Green Sand whose similarity of form to Lyidium torquilla 
 O. Schmidt, he has noticed (An. Mag. Nat. Hist. S. 4, Vol. 7, 
 p. 1 1 8, PI. VIII). Judging from the figures, these green sand 
 spicules are of a greater thickness than those from Horstead. 
 Spicules resembling those of Lyidium are figured by Mr. Wright 
 from the North of Ireland (op. cit. PI. Ill, figs. 2 a, b, 3 a, b) 
 and are referred to by Dr. Bowerbank as singular and pro- 
 bably abnormal form of dermal spiculum of a siliceo-fibrous 
 sponge I have also seen Lyidium spicules from the Chalk 
 of Oxfordshire which were discovered by the Rev d R. Pattrick 
 of Warbro' Lastly, they are figured by Zittel from the Chalk 
 Formation of Westphalia (Ueber Coelop. Taf. VII, fig. 38 
 
54 
 
 and I have myself obtained them in great abundance in ma- 
 terial from Coesfeld. 
 
 Lyidium cretacea, n. sp. 
 
 (Plate IV, figs. 1013). 
 
 Moderately robust spicules of very various forms, the 
 branches sometimes expanded at their extremities but frequently 
 elongated and obtusely pointed. The mesh work of the ske- 
 leton of a more open character than in the preceding species. 
 Average thickness of the spicules 0,09 mm. 
 
 The spicules of this species can be readily distinguished 
 from those of L. Zitteli^ by their lesser thickness, the greater 
 frequency in which the spicular arms terminate obtusely and 
 by the openings of the net work being proportionately larger 
 Not very abundant. 
 
 Genus Carterella, Zittel 1878. 
 Carterella, sp. 
 
 (Plate IV, figs. 1423). 
 
 Robust, straight or curved , elongated spicules , smooth, 
 cylindrical or compressed, having at their extremities a small, 
 thin, tongue-like prolongation frequently bent into a hook- 
 like form. Occasionally the end of the spicule is blunted 
 and forms a slight knob. The spicule is also, near the extre- 
 mities, and sometimes in the centre, transversely grooved or 
 deeply notched. As a rule, however the central portion of 
 the spicule is smooth and rounded. Average length of spi- 
 cule 1,5 mm.; average thickness 0,085 mm. Not infrequent. 
 
 The sponges of this genus are long cylindrical bodies, 
 composed of spicules, resembling those figured. These spicu- 
 les are intertwined round each other, principally at their 
 notched extremities, like the strands of a rope; the notches 
 in the spicules permitting a very close union of the ends of 
 the spicules with each other, while the central portions of 
 
55 ' 
 
 the spicules are only laterally pressed together. Though I 
 have not obtained specimens in which the spicules are attached 
 together yet the isolated spicules so closely correspond in ge- 
 neral form with the type of the genus , C. spiculigera, Roe- 
 mer, sp. from the Upper Chalk of Hanover that there can 
 hardly rest a doubt of their belonging to it. The spicules of 
 C. spiculigera are however both longer and thicker than those 
 in the Horstcad flint, and it is therefore probable that the 
 latter belong to another species. In the Hanover spicules, 
 a central canal extends along the axis of the spicule but this 
 has been obliterated in the Horstead examples. 
 
 Family Tetracladina , Zittel. 
 
 In this group of the Lithistidae the individual spicules 
 are mostly built upon a quadriaxial type and consist of four 
 arms radiating from a common centre. The extremities of 
 the arms are split up into small branches and twigs, and it 
 is by the complicated interweaving of these minutely divided 
 extremities that the spicules unite with each other to form 
 the skeletal mesh work. Besides these four-rayed spicules of 
 the mesh, probably all the sponges of this division are fur- 
 nished with a surface layer ot spicules, mostly of a very diffe- 
 rent shape to those of the mesh. In some instances these 
 surface spicules are delicate laminae of various forms, as in 
 the Genus Plinthosella ; in others they are trifid spicules, in 
 which the shaft has been reduced to a mere point and the 
 head rays have coalesced to form discoidal heads with irre- 
 gular borders, as in Racodiscula; whilst other surface spicules 
 resemble in form the compound trifid spicules of Geodia, and 
 are only distinguished from the zone spicules of that genus 
 by their much smaller size. As a rule the individual spicules 
 of many of the sponges of this family are very minute, and 
 their slender arms and delicate branched extremities have been 
 considerably eroded by the process of fossilization. 
 
_ 5 6 - 
 
 Genus Plinthosella, Zittel 1878. 
 
 Plinthosella squamosa, Zittel. 
 
 (Studien itber fossile Spongien p. 153, Taf. 10, fig. 5). 
 
 (Plate IV, figs. 3546.) 
 
 Spicules of the mesh robust, mostly four armed, but the 
 arms unequal, straight or curved and radiating from the centre 
 at different angles. The surface of the entire spicule is com- 
 pletely covered with prominent rounded wart-like projections. 
 The extremities of the spicular arms are frequently split up 
 into small twigs, which are twined round the arms of adjoining 
 spicules , and the spicular arms themselves are closely adpres- 
 sed together so that the warty projections interlock with each 
 other ; by this means a very compact skeletal mesh work is 
 formed. Length of arm of large spicule 0,75 mm. ; average 
 thickness 0,16 mm. The surface spicules of the sponge are 
 formed by delicate laminated plates of varying forms, either 
 circular, (fig. 45) ; oval (fig. 46) ; straight and narrow, (fig. 43) 
 or curved and tapering to both ends. The narrow plates are 
 about 2 mm. in length and 0,157 mm - m width, whilst the 
 circular plates have a diameter of I mm. 
 
 The mesh spicules of this sponge are mainly characterized 
 by their robust proportions, the inequality of their arms, and 
 the prominent rounded knobs with which these are furnished. 
 In some instances, the spicules are apparently formed but of 
 three rays or arms, two of which are often in the same plane 
 and at right angles to each other, other spicules have bent 
 and curved arms, some as long again as others. Most of the 
 arms terminate in two or three minute, irregular, twig-like 
 extensions which serve to attach the spicules to each other 
 to form the mesh. These twig-like extremities do occasionally 
 intertwine with the similar terminations of adjoining spicules, 
 but, as a rule , they grasp round the sides of the spicular 
 arms, and oftentimes the arms of different spicules are, for 
 
57 ' 
 
 their entire length, in close connection with each other , so 
 that the rounded projections on their surfaces are, as it were 
 morticed together. The not infrequent specimens in which the 
 spicules yet remain attached together, prove that the skeletal 
 mesh was better able to resist disintegrating influences than 
 most of these Chalk Lithistids. The surface or dermal spi- 
 cules of this sponge strikingly differ from those of most other 
 Lithistids. They consist of extremely v thin delicate plates or 
 laminae, of regular and irregular . forms, which appear to have 
 formed a covering over the exterior of the sponge, their edges 
 overlapping each other. There is no trace of any shaft as in 
 the surface discs of Discodermia, and these spicules must have 
 been held in place by the living sarcode. Neither is there 
 any indication of interior canals; so that, in this instance at 
 least, the surface spicules cannot be regarded as modifications 
 of the skeletal mesh spicules. Some of these dermal spicules 
 have the form of straight laminae, widest in the centre and 
 gradually tapering to either end. Others of about the same 
 width are curved, and the lower portion bent round in such 
 a manner that the spicule has a tendency to stand on edge, 
 as shown in fig. 44. Other spicules are circular or oval in 
 outline whilst some few are irregular in form. Of each re- 
 gular form of these laminated spicules there are numerous 
 examples in the Horstead flint, those of each form agreeing 
 also in size with each other. The relations of these diffe- 
 rently shaped laminae to . each other was apparent at the time 
 of my picking them out of the deposit , but had they not 
 been discovered in their natural position on the surface of the 
 sponge with the well-marked mesh spicules, I think it would 
 have been inpossible to have recognized in them the dermal 
 spicules of a Lithistid sponge. 
 
 The sponges to which these mesh spicules and laminated 
 dermal spicules belong, Plinthosella squamosa Zittel, are small 
 spherical bodies from 5 mm. to 25 mm. in diameter. The 
 spicules form a coarse irregular mesh work, traversed by canals 
 
which also form irregular furrows on the surface of the skeleton, 
 beneath the "dermal layer. By the careful removal of the 
 matrix, the dermal laminae can be seen in places on the 
 surface of the sponge which they appear to have completely 
 covered and the differently-shaped laminae are present in the 
 same individual sponge. A comparison with the type spe- 
 cimens show such a close agreement both of the mesh and 
 dermal spicules that there can hardly be a doubt, but that 
 these Horstead spicules belong to the same species. P. squa- 
 mosa is from the Upper Chalk of Ahlten in Hanover. 
 
 Genus Ragadinia, Zittel 1S78. 
 Ragadinia annulata n. sp. 
 
 (Plate IV, figs. 2430. Plate V. figs. 14.) 
 
 Spicules of the mesh (PI. V, figs. I 4) minute, irregularly 
 four-armed ; frequently one arm truncated and having a round 
 knob at its termination; the other arms having a single 
 prominent ring-shaped expansion a short distance from the 
 centre, but otherwise with smooth surfaces. These arms bi- 
 furcate and terminate in irregular twig-like extensions, which 
 interlock with each other to form the mesh. Length of spicular 
 arms varying from 0,27 mm. to o,74mm. ; and in thickness 
 between 0,067 mm., and 0,135 mm. Surface spicules (PL IV, 
 figs. 24 30) composed of spicular .discs, more or less lobed 
 and branched, and having a rudimentary shaft. These discs 
 vary in diameter between 0,675 mm. and 1, 26 mm. 
 
 The mesh spicules of this species are very distinctly 
 marked off from any other of the allied forms by the peculiar 
 ring-shaped swelling with which the arms are furnished. One 
 arm of the spicule extends but a short distance from the 
 central point and then expands into a prominent rounded knob ; 
 the other three arms radiate irregularly from the centre, near 
 which each has the ring-shaped swelling, and then rapidly 
 
59' 
 
 diminishes in thickness and becomes divided into twig-like 
 branches. 
 
 The surface spicules which I regard as probably belonging 
 to this species vary in form from discs with a rudely indented 
 edge, to spicules with but a small central disc, from which radiate 
 bifurcate expanded arms. In nearly all the forms there are 
 indications of a shaft, but this is reduced to a very minute 
 point. No indications of canals are exhibited in these spicules 
 but in similarly shaped spicules from the Westphalian Chalk, 
 three small canals are shown radiating from the point of the 
 shaft, and this point is frequently not in the centre of the 
 disc. There are such numerous transitional forms between 
 the spicules with but a slightly indented disc, and those in 
 which there are three main arms with bifurcated extensions, 
 that it is impossible to separate these spicules into discoid 
 and branched forms, but, on the other hand, a distinction can 
 be made between the branched surface spicules with a central 
 disc which I place in Ragadinia, and the branched surface 
 spicules without a central disc and with much narrower arms, 
 which probably belong to the Genus Racodiscula. 
 
 The sponges which serve as the type of the Genus 
 Ragadinia\ R. rimosa, Roem. sp., are platter-shaped bodies 
 with rounded edges and short base. The spicular arms of 
 the type species are very unequal in length and are ornamented 
 with warty prominences similar to those of the G. Plinthosella. 
 Professor Zittel , in his description of the genus, states that 
 there are very numerous specimens in the Upper Chalk of 
 Ahlten in Hanover which had all been referred to one species 
 R. rimosa, Roem. but could probably be divided into two or 
 three different species. On comparing the Horstead spicules 
 with mounted specimens of the spicules of R. rimosa, in the 
 Palaeontological Museum at Munich I find that they differ in 
 having but a single ring-shaped expansion on the arms and 
 in the absence of the warty projections. There is however a 
 mounted specimen from Hanover, labelled R- rimosa, which 
 
6o 
 
 differed from the other specimens of that species and closely 
 corresponded with the Horstead spicules both in the mesh and 
 surface spicules. I think that this difference is sufficient to 
 indicate that these annulated spicules belong to another 
 species which I propose to name R. annulata. 
 
 The mesh spicules of R. annulata are rare in the Horstead 
 material, whilst the surface spicules are very abundant. 
 
 Racodiscula and allied Genera. 
 
 (Plate IV, figs. 3134. Plate V, figs. 58.) 
 
 Spicules of the mesh consisting of four equal or nearly 
 equal smooth arms radiating from a common centre; the ex- 
 tremity of the arms divided up into a mass of small twig- 
 like extensions which are united with the similarly formed 
 extremities of adjacent spicules. Length of spicular arms in 
 different spicules varying between 0,27 mm. and 0,45 mm. ; and 
 between 0,045 mm - an ^ 0,067 mm. in thickness. Surface 
 spicules consisting of horizontally expanded, numerously 
 branched delicate heads with short rudimentary shafts. The 
 width of the heads varies between 0,5 17 mm. and 1, 17 mm. 
 
 The mesh spicules of Racodiscula and allied genera are 
 very minute in comparison with those of Plinthosella and the 
 slender arms have been so much eroded by fossilization that, 
 notwithstanding they are present in considerable numbers it 
 is difficult to find good specimens. In many examples the 
 spicular arms are reduced to slender threads and their termi- 
 nations only show the stumpy ends of the fibres which formed 
 the relatively large cushion-like mass by which they were 
 united together. There are several genera of sponges, whose 
 mesh-spicules, with the exception of slight variations in the 
 length of the rays and in the size of the intricate knot at 
 their point of union, are closely similar to each other. This 
 has been shown by Zittel in the spicules of the fossil sponges 
 
of the genera Phymatella, Callopegma, Turonia and Siphonia 
 (Studien lib. foss. Spong. Taf. VIII, IX). As regards Siphonia 
 also the spicular structure has been well illustrated by Professor 
 Sollas (Quart. Jour. Geo. Soc. 1877. PL XXVI). In existing 
 sponges a similar form of spicule also occurs in Discodermia, 
 Theonella and Kaliapsis. With such small differences in the 
 structure and form of the spicule, it would be very difficult 
 to determine, even were these free spicules well preserved, how 
 many of these fossil genera may be represented, and it is 
 hopeless to make an attempt to do this, when the spicules, 
 as in the present instance, have been partially destroyed. 
 
 The surface spicules (PL IV. figs. 31 34 which I regard 
 as belonging to the same sponges as the minute four armed 
 skeleton spicules, differ from those which I have placed under 
 Ragadinia, in the absence of a central laminated disc; the 
 head of the spicule being formed by three horizontally 
 extended sinuous arms which are once, and occasionally twice, 
 bifurcated. There are two well marked varieties in these 
 surface spicules; one form (figs. 31, 32) has the branches very 
 regular and even in width, with an average diameter of the 
 head of 1,1 mm.; the other form is very small and slender 
 and the terminal ends of the branches are deeply laciniated ; 
 the diameter of this form is only 0,5 1 7 mm. These surface 
 spicules are present in greater numbers and in better preser- 
 vation than the mesh spicules, but they are also of but little 
 value for determining the particular genus of sponges to 
 which they belonged. To the smaller form (fig. 34) corres- 
 pond the laciniated spicules which Mr. Carter has described 
 from the Haldon Green Sand under the name of Dactylocal- 
 y cites Vicaryi (An. Mag. N. H. Vol. 7, p. 12 3, PL VII, figs. 
 1,2). The larger form (figs. 31,32) is nearly double the size 
 of the similarly shaped spicule from the Upper Chalk of 
 Westphalia, figured by Prof. Zittel. (Ueber Coelop. Taf. VII, 
 figs. 29, 30).' 
 
62 
 
 Dermal spicules of Lithistids. 
 
 (Plate III, figs, ir, 12, 15). 
 
 Compound trifid spicules with relatively short, straight 
 or slightly curved, pointed shaft and with spreading head- 
 rays. There are three forms of these spicules, small in com- 
 parison with the trifid spicules belonging to Geodia, and 
 probably belonging to different species of sponge. The first 
 (fig. 1 1 ) has the shaft curved and the primary head rays 
 also curved and directed upwards; they bifurcate into very 
 small pointed secondary rays. An average spicule is 0,67 5 mm. 
 in length and the thickness of the shaft 0,056 mm. In the 
 second form (fig. 12) the shaft is straight and pointed, the 
 primary head rays are very long in proportion to the shaft 
 and directed forward and the secondary rays are similarly 
 long, divergent, and projecting upwards. Total length of the 
 specimen figured 0,787 mm. diameter of shaft 0,045 mm - '> 
 width across head rays 0,675 mm. This form has been found 
 also in the Irish Chalk, (Wright: op. cit. PL II, fig. 16). The 
 third form (fig. 1 5) has also a straight shaft but the rays are 
 very short and pointed, and they bifurcate so close to the 
 top of the shaft that the primary ray can scarcely be distin- 
 guished. The length of the specimen figured is 0,99 mm. 
 thickness of shaft 0,056 mm. 
 
 As regards their dimensions, these spicules correspond 
 much closer with the trifid spicules which are present as 
 surface spicules in different genera of Lithistid sponges, than 
 with the trifid zone spicules of Geodia, and it seems to be 
 very probable, that they are allied to the surface spicules of 
 Cor alii stes, O. Schmidt. 
 
 Dermal? spicule of Lithistid? 
 
 (Plate I, figs. 29, 30) 
 
 Spicule consisting of a slender, more or less elongated, 
 nearly cylindrical shaft, having at one extremity four elongated 
 
- 6 3 - - 
 
 slender tapering arms, spread out in a plane at right angles 
 to the shaft and united together near the shaft by a solid 
 siliceous membrane. The angle embraced by the four arms 
 varies between 80 and 135 in different specimens. Only 
 traces of canals are shown in the Horstead specimens but in 
 similar spicules from Coesfeld, they are present both in the 
 rays and shaft, and unite at the junction of these. I have 
 not yet met with a complete specimen, but the arms appear 
 to have been of equal length ; the longest which I have seen 
 is 0,9 mm. The longest shaft met with is 0,787 mm. with a 
 thickness of 0.056 mm. In all the specimens the connecting 
 membrane between the arms is only developed but a short 
 distance from the shaft. The spicules are somewhat rare. 
 
 No spicule at all resembling this has been met with 
 in any recent sponge and its affinities are very doubtful. 
 The number of the rays remove it from relationship with 
 Tetractinellid sponges as also with surface spicules of the 
 known Lithistids. It is possible however that it may be the 
 surface spicule of a Lithistid sponge as there is a great 
 diversity known to prevail in the outer spicules of this group. 
 This spicule was first noticed in the Chalk of the North of 
 Ireland (Wright : op. cit. p. 90, Plate III, figs. I a, b.) and 
 Dr. Bowerbank expresses the opinion that it is 'from the 
 expansile dermal system of a siliceo-fibrous sponge' . Professor 
 Zittel has also figured several examples of the same spicule 
 from the Upper Chalk Formation of Coesfeld (Ueber Coelop. 
 p. 46 PL V. fig. 4750). 
 
 Order Hexactinellidae, Oscar Schmidt. 
 
 The sponges of this order are built up of spicules com- 
 posed of six rays which form three axes crossing each other 
 at right angles. In each ray there is a distinct canal which 
 unites with those of the other rays in the centre of the 
 spicule. There is an enlargement, in many instances, in the 
 
- 64 
 
 centre of the spicule, which is either formed by a thickening 
 of the silica of each ray at the point of junction ; or by the 
 giving off of slight buttresses from each ray which unite 
 together and form as it were a framework, of the figure of 
 a hollow octahedron, to inclose the junction of the canals 
 of each ray, in the centre of the spicule. To this latter 
 structure the term lantern node has been applied. In addition 
 to these spicules of the skeleton there are in the living 
 examples of the order very minute 'flesh' spicules of wonder- 
 ful variety and beauty of form, but in the fossil sponges 
 these have not been preserved. The skeleton of the Hexac- 
 tinellid sponges is either built up of spicules which are merely 
 connected together by the sarcode or else the spicules are 
 welded together by the extremity of their rays to form a 
 connected mesh-work with regular oval or oblong interspaces. 
 The hexactinellid sponges are also provided with an exterior 
 surface layer, composed of a delicate perforated siliceous 
 membrane which, in some instances at least, can be seen to 
 be formed by a modification of spicules similar to those of 
 the interior skeleton. In some species too the sponge was 
 anchored to the sea bottom by elongated thread-like spicules 
 terminated by small anchor-shaped heads. 
 
 In this Horstead flint the hexactinellids are represented 
 by free spicules of various sizes; by small portions of the 
 spicular mesh of different species, also by fragments of the 
 surface membrane and by portions of the root- fibre and the 
 small terminal anchors. The free spicules appear to belong 
 to sponges of the group of the Lyssakina in which the 
 spicules were only connected by sarcode. Fragments of 
 the mesh-work of the group of the Dictyonina are not 
 infrequent and occur oftener than in the case of the Lithistid 
 sponges, which may be accounted for by the fact that the skeletal 
 mesh of the Hexactinellids is formed by a complete welding 
 together by silica . of the adjacent spicular arms, and not, as 
 in the Lithistids, by a mere interlocking of their extremities. 
 
- 6 5 - 
 
 An approximation to. the affinities of these fragments of the 
 skeleton may be made by a comparison of the regular or 
 irregular disposition of the mesh , its dimensions and form, 
 and by the solid or hollow characters of the central axis of 
 the spicules. Reliable indications are also furnished in some 
 instances by the surface membrane, and the connection be- 
 tween this outer membrane and the interior skeleton is occa- 
 sionally shown by fragments in which a layer of the interior 
 skeleton spicules yet remains attached to the surface membrane. 
 Unfortunately most of these hexactinellid fragments have been 
 injured by the fossilization ; the interior canals have been 
 obliterated, the small processes forming the lantern knots have 
 been worn through, and the spicular arms have been so eroded 
 that it is difficult to tell definitely whether they were origi 
 nally smooth or spinous. 
 
 Sub-order Dictyonina, Zittel. 
 
 Genus Leptophragma, Zittel 1877. 
 
 Leptophragma sp. 
 
 (Plate V, fig. 17). 
 
 The disposition of the spicules of the mesh is irregular 
 so that the interspaces are either oblong or triangular. The 
 central node of the spicules is solid and but slightly enlarged. 
 Average thickness, of the arms of the spicules 0,078 mm., 
 length of the more regular openings of the mesh 2,25 mm.; 
 width 1,23 mm. 
 
 The solid knot of the spicules and the form and arrange- 
 ment of the mesh correspond with the structure of a spe- 
 cies of Leptophragma, from the Upper Chalk of Vordorf in 
 Brunswick, but the diameter of the mesh in the Horstead 
 examples is nearly twice that of the Vordorf sponge. The 
 hexradiate form of the spicules is frequently obscured by the 
 attachment of some of the spicules in an irregular manner to 
 
66 
 
 each other, so that some spicules appear to have seven arms 
 radiating from the central node instead of the normal number. 
 The sponges of this genus are cup shaped with a thin sur- 
 face membrane which is provided with regularly disposed 
 canal openings. Fragments of the skeleton are not uncom- 
 mon and are readily distinguished from all the other hexac- 
 tinellid remains in this flint by the large size of the mesh as 
 well as by its irregular arrangement. 
 
 Leptophragma sp. 
 
 (Plate V, fig. 1 8). 
 
 Mesh work of the skeleton arranged so as to form triangu- 
 lar and quadrangular interspaces. The central knot of the 
 spicule solid and expanded to about twice the thickness of 
 the arms. Average diameter of the mesh 1,09 mm.; thick- 
 ness of the arms of spicule 0,04 mm. ; of the central node, 
 0,09 mm. 
 
 The much smaller size of the mesh work and the infla- 
 tion of the central node of the spicule distinguish these frag- 
 ments of sponge from the preceding form. 
 
 Craticularia sp. 
 
 (Plate V, fig. 19). 
 
 Mesh skeleton very regular and disposed so as to form 
 quadrangular interspaces. The central nodes are solid, and but 
 slightly larger than the spicular arms. Thickness of the arms 
 0,033 mm.; diameter of the squares or distance between the 
 nodes 0,2 mm. Rare. 
 
 In the regularity and form of the interspaces of the mesh 
 and the character of the nodes, this sponge bears a great re- 
 semblance to specimens of Craticularia from Streitberg which 
 are in the Museum at Munich. Prof. Sollas has also descri- 
 
- 6; '- 
 
 bed a somewhat similar skeletal mesh in Eubrochus clausus 
 from the Cambridge Coprolite-bed. Geo. Mag. 1876. p. 398, 
 
 PL XIV). 
 
 Cystispongia sp. 
 
 (Plate V, figs. 20, 21). 
 
 The skeletal mesh is formed of spicules with relatively 
 short arms and large sub-spherical' solid nodes. The arrange- 
 ment of the skeleton is very irregular. Distance between the 
 nodes 0,225 mm.; thickness of spicular arms 0,033 mm. ; 
 diameter of central node o, !i mm. In addition to these frag- 
 ments there are others with the spicular arms much closer 
 arranged which appear to be dermal portions of the same 
 sponge. The remains of this sponge are more abundant in 
 the Horstead flint than those of any other Hexactinellid. 
 
 The only Cretaceous sponges with which these fragments 
 can be compared are those of the Genus Cystispongia , Roemer, 
 Figures and description of Cystispongia (Cephalites) bursa, 
 Quenstedt, sp. are given by Prof. Quenstedt in his work on 
 Die Schwamme p. 492, Taf. 138, fig. 17 and I have had 
 the opportunity of comparing examples of this species with 
 the Horstead fragments. C. bursa is usually of an oval form, 
 the outer surface is composed of a very delicate lamina of 
 silica, immediately beneath which is a layer of closely arranged 
 spicular mesh work with large spherical nodes, much resemb- 
 ling that of my fig. 21. Within the oval-shaped case of 
 silica there are thin folded layers made up of spicular meshes, 
 but the solid spherical nodes are not so large in proportion 
 to the arms as in the Horstead examples. I have discovered 
 at Horstead fragments of lamina not unlike that forming the 
 exterior membrane of Cystispongia, but none of these frag- 
 ments had any portions of the spicular mesh attached, so 
 that it is doubtful whether they really belong to this genus 
 of sponge, but the resemblance of the fragments of the ske- 
 
 5* 
 
68 
 
 leton is so close that it seems highly probable that they be- 
 long to Cystispongia. There is also a remarkable resemblance 
 between these fragments of mesh and the skeleton of Asty- 
 lospongia praemorsa. Goldfuss and under the microscope there 
 was little difference to be noticed in the dimensions , form, 
 and arrangement of the spicular mesh of this Silurian sponge, 
 and the examples from the Upper Chalk. 
 
 Coscinopora sp. 
 
 (Plate V, figs. 22 24). 
 
 Portions of skeletal mesh having circular or elliptical in- 
 terspaces of unequal dimensions; composed of the usual six- 
 rayed spicules with open or < lantern nodes at their centres. 
 Diameter of some of the larger interspaces 0,45 mm. Frag- 
 ments of outer or surface layer consisting of a delicate mem- 
 brane perforated with circular openings of different sizes, the 
 larger, which appear to be disposed at regular distances from 
 each other, have a diameter of 0,38 mm. Fragments of the 
 root-fibre composed of nearly straight spicular fibres with 
 transverse connections. 
 
 These different portions of the structure of a Hexactinel- 
 lid sponge appear to correspond with the Genus Coscinopora 
 Goldfuss. The type of the genus C. infundibulijormis, Goldf. 
 (Petref. Germ. i te Theil, p. 30, Taf. 30, fig, 10) is a cup 
 shaped sponge with thin walls, which are covered by a deli- 
 cate perforate surface membrane similar to that of fig. 23. 
 The inner skeleton has spicules with lantern nodes, form- 
 ing elliptical interspaces, which however are less regular 
 in their arrangement than in the Horstead examples (fig. 22). 
 The sponge appears to have been attached to the sea bottom 
 by a root-like extension of the base, which is iormed of 
 nearly parallel fibres of silica with transverse connections at 
 frequent intervals similar to that shown in fig. 24. Coscino- 
 pora infnndibuliformis appears to be abundant in the Upper 
 
6g- - 
 
 Chalk Formation at Coesfeld in Westphalia. Good illustra- 
 tions of the structure are given in Zittel's Handbuch der 
 Palaeontologies p. 175. 
 
 Ventriculites? sp. 
 
 (Plate V, figs. 25, 26). 
 
 Skeletal mesh work composed of spicules with stout arms 
 and comparatively large open or lantern nodes, the inter- 
 spaces are nearly circular. Distance from node to node 0,33 mm. 
 Thickness of the arms 0,05 mm.; and of the central lantern 
 node 0,15 mm. Fragments of surface layer with irregularly 
 disposed circular pores. 
 
 The skeletal mesh of this form may be readily distinguish- 
 ed from the preceding by the stouter arms, the smaller and 
 nearly circular interspaces, and their more regular arrangement. 
 These differences appear sufficient to show that there are in 
 this Horstead material two different species of hexactinellid 
 sponge with the open or lantern nodes. I am, however, un- 
 certain whether these fragments properly belong to the genus 
 Ventriculites, for the mesh has a much more regular arrange- 
 ment than in any of the specimens of Ventriculites which 
 I have seen , and corresponds closer with the mesh of the 
 allied Genus Pachyteichisma, Zittel. As however the Genus 
 Ventriculites is known to occur in the Norfolk Chalk, I prefer 
 to place these fragments provisionally under it. 
 
 Sub-Order Lyssakina, Zittel. 
 
 In this division of the hexactinellid sponges the spicules 
 are not welded together by the extremities of the arms, but 
 are merely held together by the sarcode of the living animal. 
 Consequently in the fossil condition , the spicules are as a 
 rule detached from each other, though occasionally the form 
 of the sponge is still retained. 
 
70 
 Stauractinella cretacea, n. sp. 
 
 (Plate V, figs. 9, 10, loa, n). 
 
 Large free hexactinellid spicules, arms mostly straight, 
 nearly of an equal thickness throughout or very gradually 
 tapering , one axis frequently much longer than the others ; 
 the central node solid and but slightly , if at all , inflated. 
 Extreme length of spicular arms 2,18 mm. ; thickness 0,1 12 mm. 
 Abundant. 
 
 Spicules with the above characters are numerous in the 
 Horstead flint and from their dimensions and isolated condition 
 appear to belong to the sub- order Lyssakina. In one in- 
 stance however, I have met with an example (fig. 9) in which 
 two spicules are apparently connected, but as the arms of 
 one are very short , it may be merely a case of abnormal 
 growth. The majority of these spicules have one axis much 
 longer than the others, but in none of the examples are the 
 extremities of the spicular arms complete, and it is quite pos- 
 sible that they may have extended in an attenuated form to a 
 considerable length, and would have then presented a greater 
 resemblance to the spicules of the existing sponges of this 
 sub - order. One spicule (fig. 10 a) differs from the rest in 
 having one arm truncated and rounded a short distance from 
 the node, and the other arms are thinner and pointed. The 
 interior canals of these spicules are but rarely preserved, 
 when present they are of an abnormal width. 
 
 The relationship of these free hexactinellid spicules ap- 
 pears to be very near to Stauractinella jurassica, Zittel 
 (Studien iiber fossile Spong. p. 60) in which similarly shaped, 
 but much larger spicules than these from the Chalk Strata, 
 form the skeletons of large spheroidal sponges. As they are 
 readily distinguished from the Jurassic forms , I propose to 
 name the sponge to which they belonged Stauractinella cre- 
 tacea. A similar spicule, but so far as I can judge, of much 
 smaller size has been found in the Chalk of the North of 
 Ireland (Wright: op. cit. PI. Ill, fig. 4). 
 
y f 
 
 Genus Hyalostelia, Zittel. 
 
 (Handbuch der Palaeontologie p. 185). 
 
 Hyalostelia fusiformis n. sp. 
 
 (Plate V, fig. 12 16). 
 
 Free hexactinellid spicules having one axis much longer 
 than the others, arms straight or but slightly curved, and 
 gradually tapering to a point. The centre of the spicule 
 solid and inflated. Very variable in their dimensions. The 
 longer axis of the larger spicules is 1,05 mm. in length; and 
 the central node o. 1 8 mm in thickness. A small spicule is 
 0,45 mm. long; and the central node 0,09 mm. in diameter. These 
 spicules are distinguished from the preceding form by the rapid 
 tapering of the arms and the prominent inflation of the central 
 node. In some examples the arms of the spicule are fusiform, and 
 they have the aspect of three robust acerate spicules joined 
 together at right angles ; in others, the central node is more 
 of a spheroidal character and the arms are more distinctly 
 marked off from the centre. The smaller spicules are more 
 numerous than the larger in the deposit. 
 
 The forms nearest to these spicules are those which have 
 been described by Messrs. Young from the Carboniferous strata 
 in Scotland under the name of Hyalonema Smithii , (An. 
 Mag. Nat. Hist. S. 4, Vol. 20, p. 425, PI. XIV). The Hor- 
 stead specimens are however much more regular in their form, 
 as well as much smaller than the Carboniferous spicules. The 
 peculiar fusiform inflation of the central node marks these 
 spicules off very clearly from those of any known fossil and 
 recent species of this sub-order. 
 
 A spicule approximating in size to the smaller of my 
 specimens has been figured by Mr. Carter from the Haldon 
 Green Sand (An. Mag. N. H. S. 4, Vol. 7, p. 123, PI. VII, 
 fig. 15) and he compares it to the silicified fibre of the 
 Euplectellidae. A spicule with inflated node but with the 
 
arms of an even thickness throughout has also been found in 
 the Irish Chalk. (Wright: op. cit. PL III, fig. 5.) Dr. Bower- 
 bank has also delineated a spicule from the existing sponge 
 Euplectella aspergillum Owen, (Monogr. Brit. Spong. Vol. I, 
 p. 257, PL VIII, fig. 174) which somewhat resembles the 
 larger spicules from Horstead, but it is only about one-fourth 
 the length and the central inflation is not so pronounced. 
 
 Anchoring Spicules of Hexactinellids. 
 
 (Plate I, figs. 3136. Plate V, figs. 27.) 
 
 Mostly small spicules with elongated shaft and variously 
 shaped head from which 4, and in one instance 6 rays or 
 barbs project backwards. There are present in the Horstead 
 flint several different forms of these anchor spicules which are 
 all characterized by a more or less inflated head and a 
 shaft which is short in all the examples preserved, but may 
 originally have been extended. One of these forms (figs. 34, 
 36) has a very slender shaft and the head shaped like a four- 
 sided pyramid from the base of which four minute rays 
 project backwards. The shaft in some of these spicules 
 appears to have been when complete short and pointed and 
 it is doubtful if these anchor spicules really belong to Hexac- 
 tinellid sponges. Length of these spicules 0,40 mm ; average 
 width of head 0,112 mm. In another form of spicule (figs. 33, 
 35) the head has the form of a cone, from the base of which 
 four or six minute rays project backwards at an acute angle 
 with the shaft. An example of this form (fig. 33) has a 
 length of 0,607 mm. an< ^ the width of the head is O,i8mm. 
 A third form (figs. 31. 32) has a more robust shaft than the 
 others, and the head is obtusely rounded with four curved 
 arms radiating from it like the ribs of an umbrella. In one 
 specimen, the length of the incomplete spicule is 0,5 17 mm. 
 and the expansion of the head rays 0,45 mm. The shafts of 
 
73 
 
 all these spicules appear to have been smooth but there occur 
 fragmentary spicules (Plate V, fig. 27) consisting of a straight 
 nearly cylindrical shaft from the surface of which numerous 
 minute recurved spines project, which probably are the shatts 
 of anchoring spicules. Both the anchoring spicules and the 
 spinous shafts are very rare. 
 
 The resemblance of these anchoring spicules to those of 
 existing sponges of the sub-order Lyssakina is sufficient to 
 prove that they belonged to this group, and not improbably 
 to the same sponges as the free hexactinellid spicules already 
 described. A comparison may be made with the figures of 
 several kinds of anchoring spicules of existing sponges given 
 by Mr. Carter (An. Mag. Nat. Hist. S. 4, Vol. 12, PL XIV) 
 also with those of Rossella antarctica , Carter (An. Mag. 
 Nat. Hist. S. 4. Vol. 9, PL XXI). Similar shaped but larger 
 anchoring spicules also occur in Hyalostelia (Hyaloneina) 
 SmitJiii Young and Young (An. Mag. Nat. Hist. S. 4, Vol. XX, 
 PL XIV) and more recently Mr. Carter has found them fossil 
 in carboniferous strata near Sligo in Ireland (An. Mag. Nat. 
 Hist. S. 5, Vol. 6, p. 211, PL XIV, fig. 9). Mr. Wright has 
 also figured a small four armed anchoring spicule with spi- 
 nous shaft from the Irish Chalk (op. cit. PL II, fig. 23 a, b). 
 
 Spicules with Borings. 
 
 (Plate V , figs. 28, 29). 
 
 Out of several hundreds of sponge spicules from Hor- 
 stead which I have examined under the microscope, there are 
 a few which give distinct evidence of having been perforated 
 by some organism. The perforations are of the form of 
 simple, unbranched, elongated, cylindrical tubes extending for 
 various lengths in the interior of the shafts of acerate and 
 trifid spicules. The tubes are sometimes curved and smooth, 
 but occasionally they are convoluted and twisted. They ter- 
 minate blindly in the interior of the spicule. In some 
 
74 
 
 instances the tube apparently commences in the centre of the 
 broken end of a spicule, where the canal in the perfect example 
 would be situated and from thence it extends forwards in the 
 substance of the spicule without any regularity , occasionally 
 keeping near the centre or wandering towards different parts 
 of the circumference, without however penetrating the outer 
 wall of the spicule. The interior is apparently filled with a 
 transparent material of a light greenish tint, giving no colour 
 under the polariscope and apparently destitute of any granular 
 particles. The longest of the tubes noticed is 1,8 mm. and 
 the width varies in different specimens between 0,033 nim - 
 and 0,067 mm - 
 
 The mineral character of the spicules in which these bor- 
 ings occur is precisely similar to the normal ones of this 
 material, that is to say, it is how of chalcedonic silica and 
 there seems no reason to doubt that the chalcedonic silica 
 has been produced from the conversion of the original amor- 
 phous silica of the sponge spicule , so that these perforated 
 tubes whether made previous or after the change, have been 
 constructed in siliceous material. The peculiar form of the 
 perforations show that they have been produced by the action 
 of some living organism, of which no other traces are left. 
 These tubes are larger than those which have been described 
 by Prof. Duncan in the substance of Silurian and Tertiary 
 corals, and attributed by him to the boring influence of a 
 parasitic alga, PalacacJiyla perforans (Quart. Jour. Geo. Soc. 
 Vol. XXXII, p. 205) neither are they, branched or filled with 
 the dark granules which appear associated with the borings 
 in the corals. There is further the peculiarity that they have 
 been made in siliceous structures. 
 
75 
 
 On account of the detached and free condition, in which 
 most of the sponge spicules here described, are met with, it 
 is apparent that nothing more than an approximate estimate 
 can be formed of the number of species and genera which 
 are represented in the contents of this flint. A certain degree 
 of accuracy can be obtained in the case of many Lithistid 
 and Hexactinellid sponges in which the peculiar form of the 
 individual spicules and the not seldom instances in which 
 these remain together, mark off one species from another; 
 but in the case of Monactinellid and Tetractinellid sponges in 
 which only isolated spicules are met with, and are often 
 closely alike in different species, there is great difficulty in de- 
 ciding to how many species the various spicules may belong. 
 Making due allowance for this fact, I estimate that the 1 60 
 different forms of spicules which are here described and figu- 
 red from the cavity of this one chalk-flint belong to 38 spe- 
 cies and 32 genera of sponges. These are divided into 4 
 species of 3 genera of Monactinellid sponges; 2O species of 
 7 genera of Tetractinellids; 6 species of 5 genera of Lithis 
 tids ; and 8 species of 7 genera of Hexactinellids. Whatever 
 may be the opinion as to the validity of so many species 
 and genera, no doubt can remain that these spicules and frag- 
 ments of sponges which are thus mingled together in one lo- 
 cality and in so small a quantity of material, bear incontrovertible 
 evidence of a very great development of the different orders 
 
of sponges in the English Chalk. As the preservation of 
 these spicules depends so much upon their capacity to resist 
 the eroding influences of fossilization which adversely affects 
 the minuter forms far more than the larger, too much stress 
 cannot be laid upon a calculation of the relative abundance 
 of the different orders of sponges, from the spicules which 
 have escaped destruction. To some extent the great number 
 of species ot the Tetractinellid sponges, which are more nume- 
 rous in this material than those of the other orders put to- 
 gether, may be due to the fact that the zone and body spicules 
 of this order are unusually robust, and would thus be more 
 likely to be preserved than the smaller spicules of other 
 sponges. But even allowing for this, the Tetractinellid sponges 
 appear to have predominated, and next to them the Lithistids 
 and Hexactinellids are in about equal numbers, whilst the 
 Monactinellids, so far as can be determined, are few and un- 
 important. It may perhaps be deemed improbable that such 
 a number and variety of sponges should be mingled together 
 in such a small quantity of material but similar instances of 
 the occurrence of great numbers of sponges together have 
 also been discovered in modern deep-sea dredgings. Thus 
 for example, Sir W. Thomson records that in one haul of the 
 dredge in the North Atlantic there were brought up forty 
 specimens of vitreous sponges (An. Mag. Nat. Hist. 1869, 
 p. 119) and now lately, in material from the Gulf of Manaar 
 -in the Indian Ocean which in quantity would hardly fill a 
 quart measure; Mr. Carter has described no fewer than 62 
 species of sponge (An. Mag. N. H. S. 5, Vol. 6, p. 457); 
 so that the contents of this chalk-flint are to some extent pa- 
 rallelled by the deposits of the present oceans. 
 
 It is doubtful whether any precise conclusions as to the 
 depth of the ocean in which these chalk sponges existed, 
 can be drawn from comparing them with existing sponges 
 whose bathymetrical limits have been ascrtained, on account of 
 the great limits within which sponges of the same genus oftentimes 
 
77 
 
 appear to exist. At present the results of the investigations of 
 the sponges ot the Challenger expedition are not known, but 
 recent investigations of the Atlantic sponges by Mr. Carter 
 and those from the Gulf of Mexico by Oscar Schmidt show that 
 the previous ideas as to the depth in which different orders of 
 sponges exist, have to be considerably modified. The Mon- 
 actinellidae and Tetractinellidae were formerly regarded 
 as inhabiting comparatively shallow waters, but it is now 
 ascertained that they extend to depths previously thought to be 
 exclusively inhabited by Lithistidae and Hexactinellidae ; whilst, 
 on the other hand , some representatives of these latter are 
 now known to exist in but moderate depths. Two or three 
 examples may be given of the depths in which existing 
 sponges nearly allied to those in the chalk, have been dis- 
 covered. A species of Pachastrella, P. amygdaloides, Carter, 
 whose spicules most closely resemble some in the Chalk flint , 
 exists in the Atlantic at a depth of 1752 feet; Lyidium 
 torquilla. O. Schmidt to which the chalk spicules of the 
 same genus are nearly allied, also exists in the Atlantic, off 
 the coast of Cuba, at a depth of 1610 feet; Lithistid sponges 
 nearly allied to Racodiscula occur at varying depths between 
 720 feet and 1620 feet; the Hexactinellidae are most abundant 
 in depths between 1800 and 6000 feet, but a species of 
 Cystispongia a genus also represented in the Chalk flint is 
 recorded by Oscar Schmidt (Spong. d. Meerbusen von Mexico) 
 as existing at depths varying between 120 feet and 1752 feet. 
 So far as these comparisons extend, the sponges of this 
 chalk flint may have inhabited depths of 1700 feet. 
 
 The great resemblance which is presented by these 
 Horstead sponge remains and the spicules which have been 
 described from the Haldon Green Sand; the Upper Chalk of 
 the North of Ireland, the Chalk Formation of Westphalia, 
 and the Eocene Sand of Brussels shows that these sponges 
 had a wide distribution both in space and time. In the 
 Eocene strata of Brussels there occur many forms of spicules 
 
which are not found at Horstead ; but with few exceptions 
 in this single Horstead flint, there are present all the detached 
 forms of spicules which have been found both at Haldon and 
 in Ireland and Westphalia. Of the 34 different forms of 
 spicules from the North of Ireland which were obtained 
 from the contens of hollow flints from 36 different 
 localities all the forms except one are met with in 
 the single flint at Horstead. Detached sponge spicules 
 are also known from the Upper Chalk in the South 
 East of England but up to the present, so far as I am aware, 
 no published description has been given of them. The few 
 sponges described hitherto from the Upper Chalk of England 
 belong to the Lithistidae and Hexactinellidae whose more or 
 less complete skeletons occur in a few localities, but are by 
 no means generally distributed. From these sparsely distri- 
 buted specimens no one would be justified in stating that the 
 sponges played an important part in the building up of the 
 Upper Chalk of England. And even at Horstead one would 
 be extremely fortunate to find a fragment of sponge skeleton 
 in the Chalk itself. Yet in this same locality the contents 
 of this hollow flint make it manifest that sponge life was equally 
 as abundant and varied, and contributed as great a share to 
 the deposits of the cretaceous ocean as these organisms do 
 to the present deep-sea deposits in the Atlantic. The per- 
 fect state of preservation, not only of the sponge spicules, but 
 also of the tender shells of foramenifera and ostracoda which 
 in marvellous variety of form are all mingled together in the 
 cavity of this flint, proves that this flint meal contains a fail- 
 sample of the organisms which originally formed the creta- 
 ceous ooze, if we except however the coccoliths which are not 
 preserved. Subsequent metamorphism has altered this ooze, 
 filled with perfect organic remains, into the present beds of 
 Chalk and layers of flint nodules. As already noticed, the 
 organic remains in the flint meal have not altogether escaped 
 a certain degree of alteration, which has affected their mineral 
 
79 
 
 constituents, but has stopped short of destroying their forms or 
 dissolving them. A study of these fossils thus wonderfully 
 preserved enables us rightly to interpret the changes 
 which have taken place in the production of the chalk 
 and flints out of the cretaceous ooze. The foramenifera, 
 ostracoda, and other small organisms whose skeletons 
 are formed of carbonate of lime have been for the 
 most part so crushed and broken up by the mechanical 
 pressure to which they have been subjected, that though the 
 fragments are sufficient to indicate their former existence, it 
 is only here and there under special circumstances that their 
 skeletons are found complete in the chalk. A still more 
 complete change awaited the skeletons of the sponges. These 
 structures, after the death of the animal, appear largely to 
 have fallen apart into their constituent spicules, and though 
 composed of silica, which till lately was regarded as more 
 stable than carbonate of lime, they were so completely dis- 
 solved and removed from the chalk that an analysis of this 
 material in which they flourished, does not yield a trace of 
 silica to show their former presence in it. The silica resulting 
 from the solution of these sponges became aggregated to form 
 the nodules of the flints or was deposited in the joints and 
 fissures which intersect the chalk strata in many places. In 
 the same proportion as the chalk indicates the remains of the 
 foramenifera, ostracoda, and other organisms with calcareous 
 skeletons, so also do the imbedded nodules and masses of 
 flints bear witness to the siliceous skeletons of the sponges 
 which contemporaneously existed with them. This may appear 
 a somewhat far-rtaching conclusion to arrive at from the pre- 
 sence of these sponge spicules in this single chalk flint but 
 evidence from other quarters corroborates and strengthens it. 
 Mr. Wright's investigations in the material inclosed in chalk 
 flints from various places in the North of Ireland prove that 
 there also siliceous sponges abounded mingled with the shells 
 of foramenifera and other calcareous organisms, whilst in the 
 
8o 
 
 chalk itself, which has there been metamorphosed into a hard 
 limestone, no traces of sponges remain, and nearly all recog- 
 nizable fragments even of the calcareous organisms have- been 
 obliterated. It is possible that a small portion of the silica 
 of the flints may be derived from other organisms such as 
 the Radiolaria ; but up to the present the only two or three 
 forms of this group known to exist in the Chalk formation 
 have been discovered by Prof. Zittel in strata in Westphalia 
 teeming with unaltered sponge spicules ; and so far as known 
 at present the Radiolarians had but a slight development in 
 the cretaceous ocean. 
 
 The hypothesis of the derivation of the flints in the chalk 
 from the silica of sponge skeletons is by no means a new 
 one, but it has not hitherto been generally accepted from the 
 absence of proof of the existence of sponges in sufficient 
 numbers to furnish silica for the great mass of flints with 
 which the chalk is filled, and from the difficulty of explaining 
 the arrangement of the flints in nodular layers. Even in the 
 last edition of the Students Elements of Geology published 
 in 1871, Sir Charles Lyell attributed the chalk flints to the 
 dissolved shells of diatoms which at certain periods swarmed 
 in the cretaceous ocean and were alternately replaced by the 
 foramenifera which supplied the material for the chalk. Within 
 this present year the origin of the cretaceous flints has been 
 made the subject of a paper by Dr. Wallich. (Quart. Jour. 
 Geo. Soc. 1880, p. 68). This able investigator brought no 
 evidence to show the general distribution of the sponges in 
 the cretaceous ocean, but merely based his argument that the 
 flints were derived from sponge skeletons, on the fact, that 
 siliceous sponges are found abundantly in the deep sea dredg- 
 ings of the Atlantic, and that on the assumption of their 
 being present in similar numbers in the Chalk ocean, they 
 would have been able to furnish the silica for the flints. The 
 contents of this flint from Horstead and of those from the 
 North of Ireland, prove, what Dr. Wallich assumed, that in 
 
Sl- 
 its original condition the cretaceous ooze was, like that of 
 the Atlantic deep sea mud, filled with the spicular skeletons 
 of sponges. At the same time they furnish no evidence of 
 the hypothesis of Dr. Wallich that alternating periods are 
 established during which one of the two predominant animal 
 types (Foramenifera and Sponges) gradually overwhelms and 
 crushes out the other over indefinite local areas, the strata of 
 the chalk in the one case and the intercalated flint beds in 
 the other, being the issue of these contests, (op. cit. p. 72) 
 The contents both of the Irish and Horstead flints show that 
 the sponge spicules are equally as much intermingled with 
 foramenifera and other calcareous organisms as in the Atlan- 
 tic ooze, and that therefore both these animal types flourished 
 contemporaneously without alternately choking the life out 
 of each other. 
 
 These hollow flints throw no fresh light upon the causes 
 which have brought about the arrangement of the nodules in 
 definite layers but the perfect state of preservation ot their 
 contents show that in certain cases at least they were formed 
 before there was any great accumulation of overlying material, 
 which would have crushed the tender shells inclosed within. 
 But whatever may have been the forces which dissolved and 
 re-deposited the silica in its present form, it seems evident 
 that the delicate skeletons and spicules of sponges have been 
 the original source of the material itself. 
 
POSTSCRIPT. 
 
 Whilst this work was going through the Press, I received 
 from Professor Sollas, a separate copy of his paper on the Flint 
 Nodules of the Trimmingham Chalk, which appeared in the No- 
 vember and December numbers of the Annals and Magazine of 
 Natural History. In this paper, Prof. Sollas has described diffe- 
 rent forms of sponge spicules derived from material of the same 
 character, in all essential respects, as that from Horstead. So 
 far as one can judge from the outlines of the spicules in the 
 accompanying plates, they are, with few exceptions, of similar 
 forms to those which I have herein described, as also to those from 
 the Chalk of Westphalia and the North of Ireland, figured by 
 Zittel and Wright respectively. The Trimmingham nodules, how- 
 ever, on the supposition that they have been thoroughly search- 
 ed, do not contain anything like the variety of spicular forms 
 which are present in the Horstead flint. Prof. Sollas has placed 
 then under 17 genera, of which no fewer than 13 are new; 
 whereas I have placed the far greater number of spicules from 
 Horstead, including nearly all the forms present at Trimmingham, 
 under genera already known, to which they appeared closely 
 allied. As these spicules so nearly resemble those of sponges 
 already described, I am unable to see either the necessity or 
 advantage of instituting new genera to contain them, and have 
 therefore grouped under one genus, spicules, which Prof. Sollas 
 has placed under two or three genera. 
 
 The well-marked manner in which the skeleton of Lithistid 
 sponges is built up by the interlocking of the modified arms of 
 
-8 3 - 
 
 the spicules, distinguishes this group so clearly from that of the 
 Tetractinellidae that it is difficult to see why it should be rele- 
 gated to merely sub-ordinal rank, as has been done by Prof. Sollas. 
 The analogy of the Lyssakine and Dictyonine Hexactinellidae does 
 not hold good with respect to the Lithistidae and Tetractinellidae ; 
 for the Lithistid skeleton is not formed by the mere welding to- 
 gether of the usual spicules of Tetractinellid sponges, but the spic- 
 ules themselves materially differ. It is true that, not infrequently, 
 detached trifid and quadrifid spicules occur in Lithistid sponges 
 and show a near relation between the two orders ; but in the 
 same manner, simple monaxial spicules are present in all the 
 orders of siliceous sponges, without being regarded as sufficient 
 reason to group theni all as mere sub-divisions of the Monac- 
 tinellidae. 
 
 Prof. Sollas has stated (loc. cit. p. 393) that certain forms 
 figured in Prof. Zittel's Monograph on Coeloptychium (Taf. V figs. 
 n, 12, 17) as sponge spicules, are only casts of foramenifera. 
 In this he is most certainly under a mistake, for such is the 
 beautiful state of preservation of the sponge remains from the 
 Westphalian Chalk, that a novice would be able to distinguish 
 a sponge spicule from the cast of a foramenifer. 
 
PLATE I. 
 
 page. 
 Figs, j 3 Geodia and allied Genera. Different forms of acerate spi- 
 
 cules. 2 T 27 
 
 Fig. 4 Geodia clavata , n. sp. Shaft , probably of trifid spicule 
 
 showing very wide interior canal. 2 T 29 
 
 Fig- 5 Scolioraphis ? sp. 2 T 23 
 
 Figs. 6 9 Ophiraphidites sp. Different forms of curved spicules. 2 T 28 
 
 Figs. 10 75 Acuate spicules of Monactinellids. 2 T 20 
 
 Figs. 16, 17 Reniera, sp. Cylindrical curved spicules. * r 23 
 
 Figs. 18, 79 Reniera, sp. Conical spicules , with smooth outer surface 
 
 and interior canal. 2 T 21 
 
 Figs. 20, 21 Geodia and allied Genera. Small acerate spicules. j . . 27 
 
 Fig. 22 Reniera sp. Spined spicule. 6 r 21 
 
 Fig' 23 Stelletta? sp. Siliceous disc showing canals. 6 f .... 40 
 
 Fig. 24 Stelletta? sp. Siliceous disc with scalloped border. 6 r . . 40 
 
 Fig. 25 Geodia and allied Genera. Siliceous globule. 6 T ... 38 
 
 Figs. 26, 27 The same. Globostellate spicules. 6 r 38 
 
 Fig. 28 Stelletta? sp. Siliceous disc, imperfect at the edges ; showing 
 
 the canals. 6 j 40 
 
 Figs. 29, jo Probably dermal spicule of Lithistid sponge. Fig. 30 shows 
 
 the shaft. \ 62 
 
 Figs, 31, 32 Four-armed anchoring spicules of Hexactinellid sponge. 6 T 72 
 
 Fig. 33 Anchoring spicule of Hexactinellid with six rays. 6 ^ . . 72 
 
 Figs. 34 36 Four-rayed anchoring spicules of Hexactinellid sponge. ^ 72 
 
 All the figures are drawn by means of the Camera lucida. 
 
HINDE. Fossil Sponge Spicnles. 
 
 Plate I 
 
PLATE II. 
 
 page. 
 
 Fig. i Geodia? clavata n. sp. Zone spicule with bifid head" ... 29 
 
 Fig. 2 The same. Zone spicule with simple trifid head 29 
 
 Fig. S The same. Zone spicule with compound trifid head ... 29 
 
 Fig. 4 The same. Zone spicule with compound trifid head ... 29 
 
 Fig. 5 The same. Head of compound trifid spicule 29 
 
 Fig. 6 Geodia ? coronata n. sp. Zone spicule with simple trifid 
 
 head 31 
 
 Figs. 7, S The same. Zone spicules with compound trifid head ... 31 
 
 Figs, q, 10 Stelletta sp. Zone spicules with simple trifid head 33 
 
 Figs, n, i la Stelletta sp. Zone spicule with simple trifid head 33 
 
 na Head of spicule seen from below 33 
 
 Fig. 12 Geodia? Wrightii n. sp. Zone spicule with simple trifid head 31 
 
 Figs. 13 75* Geodia? sp. Simple and compound trifid spicules 35 
 
 Fig. ib Geodia? sp. Simple trifid spicule 36 
 
 Figs, i j /8 Geodia? sp. Simple trifid 'fork' spicules 35 
 
 Fig. 19 Geodia? sp. Bifid 'fork' spicule 35 
 
 All the figures on this plate are drawn to the scale of 20 diameters by means 
 of the Camera lucida. 
 
HINDE. Jossil Sponge SpicuLes 
 
 Plate II, 
 
 T3r. Efiller,! IDmcTaen. . 
 
PLATE III. 
 
 p:ige. 
 
 Figs, i , 2 Tethya ? sp. Simple trifid ,zone' spicules 36 
 
 Fig. 3 Geodia ? sp. Compound trifid zone spicule 34 
 
 Fig. 4 Tethya? sp. Simple trifid zone spicule 37 
 
 Fig. j Tethya ? sp. Simple trifid zone spicule 37 
 
 Figs. 6, 7 Geodia? sp. Simple and compound trifid spicules 34 
 
 Fig. 8 Tethya? sp. Simple trifid zone spicule 38 
 
 Fig. g Geodia? sp. Compound trifid spicule 34 
 
 Fig. 10 Tethya? sp. Simple trifid zone spicule 37 
 
 Figs, u, 12 Dermal spicules of Lithistids 62 
 
 Figs, is, 14 Tethya? sp. Simple trifid zone spicules 36 
 
 Fig. ij Dermal spicule of Lithistid sp.onge 62 
 
 Figs. 1 6 23 Tisiphonia ? sp. Compound trifid zone spicules. Figs. 17, 
 1 8 , 22 are lateral views showing variously elongated 
 shafts ; in Fig. 2 1 the termination of the shaft is rounded : 
 
 Figs. 1 6, r9, 20, 23 show the Heads of different spicules . 43 
 
 Figs. 24, 25 Pachastrella sp. Quadrifid spicules 45 
 
 Fig. 2.6 Caminus? sp. Three-rayed spicule 48 
 
 Fig. 27 Pachastrella? sp. Three rayed spicule 48 
 
 Fig. 28 Pachastrella primoeva? Zittel. Five armed spicule 46 
 
 Figs. 29 31 Pachastrella Carteri, n. sp. Simple quadrifid spicules ... 46 
 Figs. 32 33 Pachastrella primoeva? Zittei. Quadrifid spicules, arms une- 
 qual and slightly bifurcated 46 
 
 Fig. 34 The same. Five armed spicule 46 
 
 All the figures are drawn by means of the Camera lucid a and to the same 
 scale of 20 diameters. 
 
H.niDE. Fossil Sponge Spicules 
 
 Plate H 
 
PLATE IV. 
 
 page. 
 figs, i 9 Lyidium Zitteli n. sp. Detached spicules of various forms ; 
 
 in Figs. 1 , 3 , spicules are shown naturally attached together 5 1 
 
 Figs. 10 /<? Lyidium cretacea n. sp. Detached spicules 54 
 
 Fig. 13 The same. Fragment of the skeleton showing four spicules 
 
 naturally attached together 54 
 
 Figs. 14 23 Detached spicules of Carterella, sp 54 
 
 Figs. 24 jo Ragadinia annulata, n. sp. Various forms of the surface 
 
 spicules 58 
 
 Figs. 3 1 34 Racodiscula and allied Genera. Surface spicules 60 
 
 Figs- 35 4 2 Plinthosella squamosa, Zittel. Detached and combined spicules 
 
 of the skeleton. Figs. 36, 40, 42 show fragments in which 
 
 the spicules are combined together 56 
 
 Figs. 43 46 The same. Differently shaped spicules of the surface layer 
 
 of this sponge 56 
 
 All the figures on this Plate are drawn to the same scale cf 20 diameters by 
 means of the Camera lucida. 
 
HIKDE. Fossil Sponge Spicules 
 
 Plate N. 
 
 SB^art 
 
 <f /w 
 &. 
 
PLATE V. 
 
 Pag. 
 
 figs, i 4 Ragaclinia annulata n. sp. Different forms of single 
 
 and combined skeleton spicules. Fig. 3. shows two 
 
 of the spicules attached together 4 j 58 
 
 figs, j" 8 Racodiscula and allied genera. Detached and com- 
 
 bined skeleton spicules. Fig. 7 shows a portion of 
 the skeleton with the spicules naturally attached 
 
 together 4 T 60 
 
 figs. 9, 10, loa, n Stauractinella cretacea n. sp. Free skeletal spicules 70 
 figs. 12 16 Hyalostelia fusiformis. Free skeletal spicules ... 71 
 
 Fig. 17 Leptophragma, sp. Fragment of the skeletal mesh . 65 
 
 fig. 1 8 Leptophragma, sp. Skeletal mesh with solid nodes 66 
 
 fig. 79 Craticularia sp. Skeletal mesh with solid nodes . . 66 
 
 fig. 20 Cystispongia sp. Skeletal mesh with solid spherical nodes 67 
 
 fig. 21 The same. Surface layer of the skeletal mesh . . 67 
 
 fig. 22 Coscinopora, sp. Fragment of the skeletal mesh with 
 
 lantern nodes 68 
 
 fig. 23 The same. Portion of the outer membrane .... 68 
 
 fig. 24 The same. Fragment of the root-fibres 68 
 
 fig. 23 Ventriculites ? sp. Portion of the surface membrane. 69 
 
 fig. 26 The same. Fragment of skeletal mesh with lantern 
 
 nodes 69 
 
 fig. 27 Spined shaft of anchoring spicule of Hexactinellid 
 
 sponge 72 
 
 figs. 28, 29 Spicules showing irregular tubes or borings in their 
 
 interior \ 73 
 
 fig- 30 Trifid spicule, showing infiltration of the inner por- 
 
 tion of the spicule, as well as of the canal .... 14 
 
 figs, i 8, 28, 29 are drawn to the scale of 40 diameters the rest 
 of the figures are on the same scale of 20 dia- 
 meters. All are drawn by means of the Camera lucipa. 
 
HINDE. Fossil Sponge Spicules. 
 
 Hate Y. 
 
 
278894 
 
14 DAY USE 
 
 TO DESK FROM WHICH BORROWED 
 
 H SCIENCES LIBRARY 
 
 This book is due on the last date stamped below, 
 
 on the date to which renewed. 
 Renewed books are subject to immediate recall. 
 
 or 
 
 
 LD 21-40m-l,'68 
 (H7452slO)476 
 
 General Library 
 
 University of California 
 
 Berkeley 
 
U.C. BERKELEY LIBRARIES