B 
 
BERKELEY 
 
 LIBRARY 
 
 UNIVERSITY OF 
 CALIFORNIA 
 
 EARTH 
 
 SCIENCES 
 LIBRARY 
 
THE LIBRARY 
 
 OF 
 
 THE UNIVERSITY 
 OF CALIFORNIA 
 
 PRESENTED BY 
 
 PROF. CHARLES A. KOFOID AND 
 MRS. PRUDENCE W. KOFOID 
 
THE VENTRICULID.E 
 
 THE CHALK: 
 
 THEIR MICROSCOPIC STRUCTURE, AFFINITIES, 
 AND CLASSIFICATION; 
 
 INCLUDING 
 
 FIGURES AND DESCRIPTIONS OF EVERY SPECIES. 
 
 'BY 
 
 OULMIN SMITH 
 
 LONDON: 
 PRINTED BY RICHARD AND JOHN E. TAYLOR, 
 
 RED LION COURT, FLEET STREET. 
 
 1848. 
 
SCIENCES 
 LIBRARY 
 
 THE following pages first appeared in a series of papers pub- 
 lished in the ' Annals and Magazine of Natural History,' to the 
 courtesy of the proprietors of which Journal I am indebted for 
 the opportunity of reprinting them. The first of those papers 
 was published in August 1847, the last in May 1848. 
 
 The investigations carefully continued from the time of the 
 publication of the first of those papers have served only to con- 
 firm, and greatly to strengthen the evidence of the truth of, the 
 conclusions therein expressed. Should these investigations have 
 served to open up the true reading of one page in an obscure, 
 but most interesting, chapter of the great Book of Nature, the 
 time and labour given to them will not have been mis-spent. 
 
 I have had much pleasure in acknowledging, at pages 10, 42, 
 &c., my obligations to various gentlemen for their kind assist- 
 ance. 
 
 An explanation of the Plates, fulfilling also the purpose of an 
 INDEX, will be found at the end of the volume. 
 
 J. T. S. 
 
 Highgatc, 20th May, 1848. 
 
 M375125 
 
[From the ANNALS AND MAGAZINE OF NATURAL HISTORY /or July 1848.] 
 
 THE remarks of Dr. Mantell in the ( Annals of Natural History ' 
 of the present month call for a few brief observations from me. 
 I must always regret a difference of views from any one for whom 
 I have a respect, and especially with whom I have been in 
 habits of friendly intercourse, when, but only when, any feel- 
 ing of reserve or asperity arises out of that difference of views 
 instead of the cordial desire for still further discussion and in- 
 vestigation, and so getting, in the end, nearer to the truth. 
 
 Dr. Mantell imputes to me presumption in stating that " the 
 field was an entirely untrodden one and the task a new one," 
 and that " the nature of this class of animals was totally un- 
 known before." I wish to make two answers to this imputation ; 
 first, as regards an implied denial of justice to himself; second, 
 as to the matter of fact. 
 
 First, as to my notice of Dr. Mantell himself. In pp. 1 and 2 
 I use the following language. Having named Dr. MantelFs 
 paper in the ' Linnsean Transactions ' I add : " That paper was 
 but one among the many results of the indefatigable labours of 
 its author in a field then little trodden and by few feet. * * It 
 can be no reflection on the Discoverer of the Wealden and First 
 Investigator of the Chalk to show that, amid the multitude of 
 objects which engaged his attention, one was not followed out 
 exhaustively." Again, on entering on the classification, I use, at 
 p. 50, the following words : " I have been unwilling, out of re- 
 spect for the many labours of Dr. Mantell in the field of palaeon- 
 tology, to reject, as others have done without assigning any rea- 
 
son, this generic appellation ; * * I am glad that a modification 
 in the meaning of the word enables me to retain a name which 
 will always bring to the inquirer's recollection the long and suc- 
 cessful labours of Dr. Mantell." The candid reader may judge 
 from these and other passages whether I have wished to de- 
 preciate the labours of Dr. Mantell or his reputation. 
 
 Second, as to the matter of fact. The question simply is, what 
 is knowledge of a class of animals. I apprehend that " know- 
 ledge " of any creature is not merely the sight, or bare handling, 
 or even giving an arbitrary name to a specimen ; it must imply 
 some knowledge of structure and functions, in the same way as 
 Professor Owen justly tells us that " the knowledge of the or- 
 ganized beings now called Polypi, as members of the animal king- 
 dom, is of comparatively recent introduction " (Comp. Anat. i. 
 p. 81), though these had been seen and handled on the sea-shore 
 for ages. I certainly must repeat the assertion that, previously 
 to the publication of my papers, papers as to which Dr. Mantell 
 himself has been pleased to say that " the subject has recently 
 been investigated by a gentleman of distinguished ability" 
 (Wonders of Geology, 6th ed. p. 638), the nature of these 
 beautiful fossils was " totally unknown" Neither Dr. Mantell 
 nor any previous writer had ever even suspected the existence of 
 any membrane whatever in the Ventriculidse. On the other 
 hand, < I have demonstrated that the basis of the Ventriculidse 
 is a simple unperforated membrane; that, therefore, the descrip- 
 tions so long before the world, and so often repeated, are funda- 
 mentally erroneous, the conclusions as to the oeconomy of the 
 animal being necessarily, therefore, as fundamentally erroneous " 
 (p. 48 note). In addition to this, and other points to which I 
 cannot now allude, I have discovered and described and figured, as 
 existing in this membrane, an entirely new form of animal struc- 
 ture; of which Dr. Mantell himself has said (somewhat in incon- 
 sistency with what he now considers as due to his courteous read- 
 ers) : " Of the accuracy of Mr. Toulmin Smith's beautiful mi- 
 croscopic examination of the intimate tissue of these zoophytes I 
 have, no doubt', and will only remark that the octahedral form, 
 represented as that assumed by the inosculating fibres of the 
 membrane of the Ventriculida3, is a very extraordinary anomaly in 
 animal structures" (Wonders of Geology, 6th ed. p. 638 note.) I 
 think Dr. MantelPs own words, addressed to the very readers 
 whom he now twice assures of his own accuracy, and whom he is 
 so anxious to treat with all due courtesy, are the best testimony 
 that he is doing me injustice. 
 
 I am surprised that Dr. Mantell should speak of an " incon- 
 gruous assemblage " of forms. I might take almost any family 
 of either animal or vegetable kingdom to show the futility of 
 
Ill 
 
 this objection. The Acacias platyptera, pendula and grandiflora 
 are, for instance, at first sight, almost as incongruous as are 
 Tubulipora patina, Gemellaria loriculata and Halodactylus dia- 
 phanus. 
 
 I doubt not that when the first priest of old nailed to the 
 column's head, after the solemn sacrifice was done, the scalp 
 which he had torn off the devoted ram, he thought he knew very 
 well what a ram's head was, and would have pitied any unhappy 
 wight who might have suggested any resemblance between the 
 head and the tail of the beast which had just smoked upon the 
 altar. It has been reserved for modern science not only to sug- 
 gest but to demonstrate, by one of the most beautiful, most 
 logical, most philosophical, and at the same time most scienti- 
 fically important trains of investigation that has been ever fol- 
 lowed up, that between these so " incongruous " parts there are 
 clear and positive homologies, and that no one can truly " know " 
 either part who does not study those homologies. It has been by 
 treading, though at an humble distance, and, I am fully con- 
 scious, with a too faltering step, in the path by which those im- 
 portant truths have been obtained that I have arrived at what 
 Dr. Mantell thinks fit to term " sublime transcendentalisms " 
 but which, to my mind, constitute, in every branch of science, 
 the main charm, and the most important end, of the pursuit. 
 
 In conclusion I may be allowed to say, that it is of little im- 
 portance to the world what may be Dr. Mantell' s opinion or my 
 own on the present matter ; but it is of importance w r hether, in 
 pursuing the subject, any fragment of truth has been got at. I 
 had hoped Dr. Mantell would have discussed my facts arid argu- 
 ments, and not "replied" to my conclusions. I hope that he 
 will do so yet. It is only by full discussion, close examination, 
 and careful consideration that the truth or falsity of my conclu- 
 sions can be tested. Such discussion, examination and consi- 
 deration I shall be ready and most glad to meet from any one in 
 a fair and cordial spirit. But all hope of truth and all scientific 
 investigation is at an end, if it is to be considered as a " reply" to a 
 long and most carefully conducted train of investigation that, some 
 years before, one or two of the objects whose natural history and 
 relations are thus elucidated had been described in quite a differ- 
 ent way ; and that, therefore, for that is Dr. MantelFs only ar- 
 gument, the more recent investigations must be all wrong. I 
 am glad to say that my collection has been already visited by 
 several of the most eminent paleontologists and anatomists of 
 this country ; and I know that some of these, who have the most 
 carefully examined the series, are satisfied of the truth of my 
 conclusions. Of the opinion of others I have perhaps no certain 
 information at present ; but I will only add, that, to these or any 
 
IV 
 
 others really wishing to compare the facts or to examine into the 
 nature of this most beautiful and interesting class of fossils, the 
 contents of my cabinet will, after August of the present year, be 
 at all times most freely open for inspection. 
 
 J. TOULMIN SMITH. 
 
 Higbgate, 16th June 1818. 
 
ON THE 
 
 VENTRICULIM OF THE CHALK, 
 
 ALMOST one hundred years have now elapsed since Guettard 
 drew attention to " some fossil bodies little known," the eluci- 
 dation of which he attempted in an elaborate paper*. Of the 
 two classes of bodies described by him it is clear that the 
 Choanites are one. The figures and descriptions appear con- 
 clusive on this point, and the true affinities are very shrewdly 
 pointed out by the writer, while the prevailing notion of the 
 bodies described being petrified figs and other fruits is com- 
 pletely disposed of. It may not perhaps be quite so clear that 
 the other class of fossil bodies described by Guettard comprises 
 some of the forms of the Ventriculidse. The true characters of 
 the Ventriculidse will be presently seen to be in almost all cases so 
 obscured from the general observer, and even, without careful at- 
 tention to the mode of observation, from the experienced palaeon- 
 tologist, that we cannot expect to find in either the figures or de- 
 scription of a century ago positive evidence of identity. Still I 
 think those of Guettard warrant the conclusion that objects of 
 this class were before him. 
 
 It was not till Dr. Mantell in 1814 figured and described in 
 the ' Linnsean Transactions/ vol. xi. p. 401, " a fossil Alcyonium 
 from near Lewes," that any particular attention appears to have 
 been given in this country, or, since the time of Guettard, in any 
 other, to these bodies. That paper was but one among the many 
 
 * Mem. de PAcad. Royale de Sciences for 1751. The paper is erro- 
 neously cited by Parkinson under the year 1757. I have found no other 
 direct notice of it. Michelin, in his ' Iconographie Zoophytologique,' p. 121, 
 cites M. Guettard's ' Memoires Academiques.' I have been unable to ob- 
 tain that work, but conceive it to be merely a reprint of M. Guettard's va- 
 rious scientific papers, including the one above named. 
 
results of the indefatigable labours of its author in a field then 
 little trodden and by few feet. The views expressed in it seem 
 to have undergone little modification since ; for though, appa- 
 rently rather in deference to the opinions of others than from 
 any conviction of his own, changes of opinion on some points 
 have been expressed by Dr. Mantell in later works, they are ex- 
 pressed without any grounds being stated whereon they were 
 adopted*. It can be no reflection on the Discoverer of the 
 Wealden and First Investigator of the Chalk to show that, amid 
 the multitude of objects which engaged his attention, one was 
 not followed out exhaustively. It is sufficient at present to say 
 that, whatever Dr. Mantell may have left undone in reference to 
 the Ventriculidae, has been hitherto filled up by no other hand. 
 
 The different members of the large family of Zoophytes have 
 done so much towards the actual formation of the solid crust of 
 the earth, that anything which relates to any branch of it must 
 be interesting and important ; and it is certainly remarkable that, 
 amid the great attention given within recent years by so many 
 eminent observers to this family, no one has entered on an in- 
 vestigation of the Ventriculidae And yet the wide development 
 of these forms and their great elegance and variety cannot but 
 have attracted the attention of all who have ever glanced at the 
 contents of any good and extensive series of chalk fossils ; an 
 elegance and variety which the most untutored eye cannot fail 
 to notice and admire. 
 
 I doubt not that the difficulty of the investigation affords the 
 real explanation of the neglect to which these bodies have been 
 subject. To investigate the living structure and affinities of an 
 entirely soft-bodied animal whose only remains have come down 
 to us encased either in intractable flint or in friable chalk will be 
 at once felt to be a task of no ordinary difficulty. Such a task 
 is very different from the examination of any living forms, to 
 whatever class they may belong, or even from that of any of the 
 more solid fossil forms, which latter themselves however call into 
 activity the greatest patience and skill of the ablest observers. 
 
 Space will not permit me to examine in detail all the notices 
 which have incidentally been taken of different forms of the 
 Ventriculidse. I must confine myself to a brief glance at the 
 different works in which any direct allusion to these bodies may 
 be found, leaving it to the full details hereafter to be given of 
 
 * 1 allude particularly to the separation of " Ocellaria " in the ' Medals 
 of Creation,' p. 279; to the stating Ventriculites to be a composite instead 
 of single animal in note to p. 272 of the same work; and to the still more 
 remarkable entire separation of " V. quadrangular is " (p. 283 ib>) and 
 placing it among Flustrte. 
 
the results of my own investigations to show wherein the figures 
 and descriptions in which such allusions consist have erred or 
 fallen short. 
 
 To the paper of Guettard I have already alluded. It is suffi- 
 cient further to say, that he was not content with that mere 
 superficial glance which most later observers have given. He 
 expressly says* that, though at first he regarded these bodies as 
 related to the sponges, he was obliged to abandon that idea when 
 he had given a more careful attention to the examination of 
 them ; and he concluded that their nearest affinity was to the 
 Madrepores. 
 
 In ' Edvardi Luidii Ichnographia ' (1760), tab. 2. fig. 176, is 
 figured a Ventriculite, which he describes (p. 10) as " AstroitaB 
 congener Radularia cretacea." 
 
 Parkinson, in his ' Organic Remains of a former World' (1807), 
 alludes to the paper of Guettard, and gives descriptions and figures 
 of several fossils which he considers as allied to Alcyonia, but 
 whose differences from which he yet felt to be marked. Plate 9. 
 (of vol. ii.) figs. 2, 6, 9 and 10; pi. 10. figs. 12, 14, 15 and 16 ; 
 and pi. 12. fig. 9 I consider to be certainly forms of Ventriculidse ; 
 and I think it probable that pi. 11. figs. 1 and 6, and pi. 12. 
 fig. 8, are so alsof- 
 
 In Mr. Parkinson's later ' Outlines ' (8vo, 1822) he makes 
 some sound observations on the necessity for separating from 
 the Alcyonia the various bodies figured and described in his 
 former work, but he gives no additional details of importance. 
 
 On p. 54 of the ' Outlines/ Mr. Parkinson, alluding to what 
 are undoubtedly true Ventriculidse, describes them by characters 
 which are purely external, and treats as generic characters those 
 which are merely accidental and non-essential. 
 
 The ' Organic Remains ' of Parkinson did not in the least 
 degree forestall the labours of Dr. Mantell, whose figures and 
 descriptions (1814) convey far more information on the subject 
 than all else that has even yet been published. It is but justice 
 therefore to extract at some length the description given by him 
 of these bodies. 
 
 The generic characters assigned by Dr. Mantell, as corrected in 
 his very valuable work on 'The Fossils of the South Downs' (1822), 
 p. 168, are : "Body inversely conical, concave, (1) capable of con- 
 traction and expansion : original substance spongeous ? or gela- 
 
 * Mem. p. 259. 
 
 f The frontispiece to the second volume, which is mentioned by Mr. Rose 
 (citation below, p. 339) as " so beautifully delineating " the structure of the 
 Ventriculite, is no Ventriculite at all, but an exceedingly different fossil iu 
 all respects, viz. a Wiltshire sponge. 
 
 A2 
 
4 
 
 tinous ?* : (2) external surface reticulated^ : (2) internal surface 
 covered with openings or perforated papilla : base imperforate, 
 (1) prolonged into a stirps, and (1) attached to other bodies. 33 
 
 Of the characters thus described as generic I shall hereafter 
 show that all those which I have put in italics are erroneous ; 
 those marked (1) being altogether in opposition to the fact ; 
 those marked (2) being characters which are merely accidental 
 and non-essential. 
 
 In describing these bodies in the above-cited paper Dr. Man- 
 tell says: "The specimens which occur at Lewes, though gene- 
 rally considered as Alcyonia, do not entirely conform to the cha- 
 racters of that genus as given by modern writers ; yet they are 
 evidently very nearly allied to it. It is certain that the recent 
 animal possessed great powers of contraction and expansion which 
 enabled it to assume various dissimilar forms. In a quiescent 
 state it was more or less funnel-like ; when partly expanded cya- 
 thiform ; and when completely dilated it presented the figure 
 of a broad circular disc. To this versatility of shape is to be at- 
 tributed the great diversity of appearance observable in its reli- 
 quiae, whose forms must have been derived from the contracted 
 or expanded state of the original at the period of its introduction 
 into the mineral kingdom. That the animal enjoyed the power 
 of contraction and expansion above ascribed to it, will appear 
 evident from an investigation of its structure. The epidermis or 
 external coat is composed of fasciculi of muscular fibres, which, 
 arising from the pedicle, proceed in a radiated manner toward the 
 circumference, and, by frequently anastomosing, constitute a reti- 
 form plexus capable of dilating, lengthening and contracting, 
 according to the impressions it received J. The fasciculi are further 
 connected by lateral processes , which increase the firmness and 
 coherence of the external integument. From the inner surface 
 
 * This language and query show that the author had found no specimen 
 which enabled him to ascertain the actual internal structure. 
 
 f The " reticulation " here meant was merely that of the " anastomosing 
 tubuli" named by Parkinson (8vo, p. 54), and has no reference whatever 
 to the beautiful reticulated fibrous structure hereafter to be described, and 
 which appears never to have come under Dr. Mantell's notice. If proof of 
 this remark were wanted, it is found as well in the extract which follows 
 from the paper in the Linnaean Transactions ' as in the specific description 
 given by him in the same page (South Downs, p. 168), where he speaks 
 of the " external integument composed of cylindrical, anastomosing fibres, 
 radiating from the centre to the circumference." 
 
 I Mr. Parkinson had already expressed a similar opinion as to some of 
 the fossils above named as figured by him. See ' Organic Remains/ vol. ii. 
 p. 145. 
 
 These lateral processes are in reality the fibres going off to the polyp- 
 skin hereafter described. 
 
5 
 
 of the muscular envelopment arise innumerable tubuli, which 
 pass direct to the ventricular cavity, and terminate in openings 
 on its surface. In some specimens a substance of a sponge-like 
 appearance fills up the interstices between the tubuli, and pro- 
 bably is the remains of a membrane which served in the recent 
 animal to connect the tubes and assist in strengthening and 
 uniting the whole mass. The sides of the ventricular cavity are 
 generally about one-third of an inch in thickness. From the 
 bases or pedicle proceed fibres by which the animal was attached 
 to its appropriate habitation." 
 
 The portions of the structure here described as internal, and 
 considered by Dr. Mantell to be absorbents, are in the ' Wonders ' 
 and ' Medals ; of Dr. Mantell considered as polyp cells, and the 
 animal described as a composite and not a single one. I shall 
 show that each appropriation of those so-called tubuli is erro- 
 neous. 
 
 The paper in the ' Linnsean Transactions ' is accompanied, as 
 is the description in the ' South Downs/ by many figures, the 
 truthfulness of which, as conveying the general characters of 
 outward form, has never been even approached by any later 
 writer. 
 
 M. Ramond in his ( Voyage au Mont Perdu' (before 1815 ; but 
 this is the only case in which I have been unable to obtain ac- 
 cess to the original work) figured the silicified remains of one 
 species and the cast of the same under the two names of Ocel- 
 laria inclusa and Ocellaria nuda, assigning to them characters 
 which a very little study of the nature of flint and of the process 
 of fossilization would have prevented. 
 
 In the first edition of Lamarck's ' Animaux sans Vertebres/ 
 (1816, vol. ii. p. 187), the bodies thus figured and described by 
 Kamond are included. In the second edition of that work (1836) 
 Milne-Edwards expresses doubts (p. 291) of the correctness of the 
 description which had been given of the Oceltaria, but without 
 affording any fresh insight into the real structure of the fossils. 
 In the same edition are included (p. 459) three species of what 
 Goldfuss had previously named Coscinopora, and which is a form 
 of the Ventriculidse. 
 
 William Smith, the " Father of Geology/' in his < Strata Iden- 
 tified' (1816), figures two Ventriculidse in flint (tab. 3. figs. 1 
 and 2), of which the first is a very characteristic figure. He calls 
 them Alcyonia. 
 
 In the ' Icones Fossilium Sectiles' of Konig (1820), Ramon d's 
 figures are copied (pi. 8. figs. 98 and 99), but without any 
 description. 
 
 In the ' Exposition Methodique des Genres de 1'ordre des Po- 
 lypiers ' of Lamouroux (1821) the so-called Occllaria are also 
 
6 
 
 figured and described, p. 45. pi. 72. figs. 1, 2, 3, 4, 5, but with 
 no fresh information. 
 
 The Ventriculidse were justly considered by Conybeare and 
 Phillips so interesting and remarkable that a larger proportionate 
 space is devoted to their description in the well-known ' Outlines 
 of Geology' (1822) than to any other fossil, and they are the 
 only fossils figured throughout their volume (p. 76). They only 
 however abridge the description given in Dr. MantelFs paper, 
 suggesting, however, that they were composite instead of single 
 animals as described by the latter. 
 
 In Goldfuss' s ' Petrefacta ' (1826) there are given, under still 
 new names, figures which appear to represent some of the Ven- 
 triculidse. I think it quite clear however that the two forms 
 which alone are referred to in Mr. Morris's ' Catalogue ' as figured 
 by Goldfuss (quadratus and radiatus, pi. 33. fig. 1, and 65. 
 fig. 7) are not figures of any of the family of Ventriculidse. 
 Goldfuss himself (p. 243) refers the genus Ventriculites to his 
 genus Scyphia, though it is clear one of his Coscinoporae is a Ven- 
 triculite also. 
 
 It appears to me that the following figures in Goldfuss repre- 
 sent forms of Ventriculidse ; but there is nothing in the descrip- 
 tions which enables us to identify them : tab. 2. figs. 8, 9, 10, 1 1, 
 12, and perhaps 15 and 16 ; tab. 3. figs. 1 and 5 ; tab. 30. fig. 10; 
 and perhaps tab. 32. figs. 3 and 8. It is to be observed however 
 with respect to all these figures, that they are too imperfect to 
 enable me to speak with absolute confidence of any one. It is 
 certain that he has no figure of any one of the most characteristic 
 forms of Ventriculidse. And this is not the less the case though 
 he professes to give magnified views of some of the structure ; 
 those magnified views themselves exhibiting, without exception, 
 want of accurate observation, and so being calculated to mislead 
 rather than aid the inquirer. 
 
 Mr. Rose published in vol. ii. of the ' Mag. of Nat. Hist/ (1829) 
 a paper " On the Anatomy of the Ventriculites of Mantell," in 
 which he professes to detail the intimate anatomical structure of 
 those fossils, and accompanies his descriptions with figures. The 
 figures however, which are in wood, are not such as to convey 
 any correct or clear idea of the originals, while the whole paper 
 certainly does not elucidate the structure further than had been 
 done by Dr. Mantell*. The writer considers them single animals 
 like the Actinia. 
 
 * I had proposed briefly pointing out the cause of the essential errors into 
 which Mr. Rose has fallen, but rny limits prevent (see note ante, p. 75). The 
 course which I subsequently show to be absolutely necessary to the investi- 
 gation of these bodies is the best explanation of the imperfect and erroneous 
 notions hitherto prevailing in regard to them. 
 
In Miss Bennett's ' Catalogue of Wiltshire Organic Remains ' 
 (1831) are given, but again without any description, the best 
 figures yet published of one form of the Ventriculidse, which is 
 there called. Choanites subrotundus (tab. 16. figs. 1 and 2: figs. 3, 
 4 and 5 are bad). 
 
 In Woodward's ' Geology of Norfolk ' (1833) two figures are 
 given (tab. 4. figs. 20, 21) of what the author calls Ventriculites 
 infundibuliformis, but unaccompanied by any description, and the 
 figures are too indefinite to afford any information : the author 
 includes V. radiatus in his list (p. 46). 
 
 Blainville, in his ' Manuel d' Ac tinologie ' (1834), figures on 
 pi. 76. figs. 4 and 4 a the Ocellaria of Ramond, and on pi. 60. 
 fig. 5 the Coscinopora of Goldfuss ; but in his description of each, 
 pp. 386, 430, he intimates doubts as to their real nature. He 
 describes each, however, as having a stony polypidom ! 
 
 Phillips, in his ' Illustrations of the Geology of Yorkshire ' 
 (1835), vol. i. p. 118, gives several figures, unaccompanied how- 
 ever by any description, and which figures are so very imperfect 
 that it is impossible to make out from them any character at all. 
 It would indeed be impossible to know that any of them repre- 
 sented Ventriculidse, did not the heading of the list of figures 
 state such an intention. This imperfection of these plates is the 
 more to be regretted, and the absence of all description the more 
 surprising, inasmuch as the able author himself remarks (p. 121), 
 that " the remains of the [so-called] Spongut* are nowhere so 
 well-developed as in England, and perhaps nowhere in England so 
 well as in Yorkshire. On the shore near Bridlington they lie expo- 
 sed in the cliffs and scars, and, being seldom inclosed in flint, allow 
 their organization to be studied with the greatest advantage." 
 
 Bronn, in his ( Lethsea Geognostica/ (1835-7) allows a place 
 to two of the Ventriculidse under that name, and figures another 
 under Goldfuss's name of Coscinopora (tab. 29. fig. 1). He 
 figures Goldfuss's Scyphia Oeynhausii as V. radiatus (tab. 27. 
 fig. 18), in which he is clearly mistaken. Neither the figure of 
 natural size nor magnified has any resemblance to any of the 
 Ventriculidse. 
 
 In ' Die Versteinerungen des Norddeutschen Kreidegebirges ' 
 of Roemer (1840) are figured, with very meagre descriptions, 
 some forms which seem intended to represent some of the Ven- 
 triculidse. They are however too indefinite to enable me to fix 
 
 * Though thus called " SpongUB" by this author, and though some other 
 writers have so called them also, it is really needless to expend one line in 
 showing the total absence in them of all resemblance to sponges. No two 
 classes of objects in natural history can be more different, and the affixing 
 of such a name can only arise from an entire want of opportunity for the 
 examination of specimens. 
 
8 
 
 with confidence on particular figures. All are described under 
 the name of Scyphia. The nearest forms are perhaps tab. 3. 
 figs. 2, 9, 11, and tab. 4. fig. 1, but others are probably intended 
 for these objects. 
 
 In Portlock's ' Report on the Geology of Londonderry ' (1843) 
 are contained descriptions, but no figures, of Ventriculites radiatus 
 of Mantell, and Scyphia alternans of Roemer (tab. 3. fig. 9), and 
 also of the one figured by Goldfuss (pi. 30. fig. 10) as Coscino- 
 pora infundibuliformiSf but this writer does no more in effect 
 than repeat the descriptions given by former authors. 
 
 Michelin, in his ' Iconographie Zoophytologique' (1843-7), has 
 figured (pi. 30.) and described (p. 121) under the name of Guet- 
 tardia, a variety of the Ventriculite already figured by Dr. Man- 
 tell (South Downs, tab. 15. fig. 6) under the far more character- 
 istic name of V. quadrangular is. 
 
 On pi. 38. fig. 3 of the same work is a very imperfect figure* 
 of Ventriculites Bennettice. On pi. 41. fig. 3 is a far better figure 
 than had before been given of the so-called " Ocellaria nuda '" 
 while on pi. 40. figs. 3 a and 3 b are figures of what he calls Ocel- 
 laria grandipora, being really a very different species of Ventri- 
 culite from the other so-called Ocellarice. These figures admirably 
 represent the original as it appears when first broken out of the 
 flint. The description (p. 145) contains however, in this as 
 in other cases, nothing new. On pi. 40. fig. 4 , 4 b, is also re- 
 presented, under the name of Retepora crassa, another form of 
 Ventriculite. 
 
 In ' Die Versteinerungen der Bb'hmischen Kreideformation ' 
 of Reuss (1846) no new details are given, while the figures 
 (tab. 17. fig. 14, and tab. 18. fig. 11) are remarkable, in a work 
 marked by the general beauty and correctness of its figures, for 
 the want of any character or truthfulness whatever. 
 
 Such are the notices of this very interesting class of bodies 
 which I have met with. Doubtless others may exist in works 
 which have not fallen into my hands. The above will satisfy 
 every reader that all that has been done by recent palasontologists 
 has been to copy from one another. It is important to observe 
 that in none of the figures or descriptions which I have cited 
 does there exist the slightest indication of what I shall show to 
 be the actual structure of these remarkable bodies, and without 
 an insight into which all attempts at classifying them and deter- 
 mining their affinities must necessarily be uncertain and unsa- 
 tisfactory and a true knowledge of their natural history impos- 
 sible. Dr. Mantell is, indeed, the only author who has presented 
 
 * Probably a mere copy, for Miehelin's plates are usually very good and 
 characteristic. 
 
us with an extensive series of figures and descriptions. And 
 although, as he himself properly remarks*, " some respectable 
 writers have amused themselves with either giving them new 
 names or arranging them as Spongice, Alcyonia" &c., those 
 writers have assuredly done nothing more ; and thereby, instead 
 of advancing the knowledge of these fossils, they have introduced 
 all the confusion and uncertainty which it has been possible to 
 do as to even their identity. 
 
 Dr. Johnston, in his ' British Zoophytes' (2nd ed. 1847, p. 180), 
 merely takes a passing notice of the Ventriculidse ; but his work 
 does not profess to include fossil species. 
 
 Finally, Mr. Morris, whose wand has in many instances re- 
 stored order where numberless writers had " amused themselves 
 with giving new names/' while he properly readmits the Ventri- 
 culidse to their position as a separate genus, has placed them 
 among the Amorphozoa. It is needless to dwell on the impro- 
 priety of that position, as Mr. Morris is now fully satisfied, from 
 an inspection of my collection, that they deserve a very different 
 place. And if constancy and elegance of form, delicacy of struc- 
 ture, and a high state of organization are to be taken as tests, 
 the Ventriculidse will assuredly have hereafter to be ranged in a 
 very different group from that of the Amorphozoa. 
 
 Having thus shown what has been already done in the field 
 upon which I have entered, it is necessaiy, in order that the 
 reader may have any confidence in the results and observations 
 which will be presented to his notice here, to state the course 
 which I have myself pursued in these investigations. 
 
 The first specimen of the Ventriculidse which came under my 
 notice strongly attracted my attention from its great elegance of 
 form and the peculiarities I observed in it. I was disappointed 
 in finding any satisfactory information on its nature, and soon 
 perceived that every one of the characters described in the books 
 had reference to some superficial characters only, and not to the 
 intimate structure of the animal. Being fully satisfied that " it 
 is only by a strict investigation of the intimate structure of the 
 various forms of these animals/' as an accurate observer has well 
 remarked f, "that any permanent arrangement that shall indi- 
 cate their true and natural affinities may be hoped for," I set 
 about that task myself : this has now engaged my attention for 
 upwards of two years ; and in now publishing the results of my 
 careful observations, I feel that I may add with even more truth 
 than could be done by that writer, that the task I have under- 
 taken " is a task of no little difficulty in the accomplishment, 
 
 * Geology S.-E. England, p. 97. 
 
 f Farre on the Ciliobrachiata, Phil. Trans. 1837, p. 387. 
 
10 
 
 and one that may fairly entitle him who enters upon it to expect 
 to meet with indulgence," an indulgence, however, which those 
 best qualified to judge of the value of such observations will, 
 happily, be also most able and willing to yield*. 
 
 My first step was to obtain as large and varied a series as 
 possible. I have in truth examined and compared with laborious 
 care several thousand specimens, of which certainly upwards of 
 one thousand are at this time in my own collection. These spe- 
 cimens are in all conditions and from various localities. 
 
 But I soon felt a new difficulty. These bodies exist both in 
 the chalk and in the flint, substances about as different as well 
 may be. I saw that much error and inconsistency had arisen 
 from not comparing and regarding the differing conditions in 
 which these fossils exist in such different substances. Hence 
 a preliminary step seemed to be, an examination of the nature 
 and mode of formation of the flints themselves. The conclusions 
 resulting from that examination have, to a great extent, been com- 
 municated in two papers in the ' Ann. and Mag. of Nat. Hist/ for 
 January and May of the present year. 
 
 The exposition given by Dr. Turner f of the origin of the sili- 
 ceous fluid, in the solution of silex consequent on the disinte- 
 gration of the felspar of igneous rocks, at which moment of dis- 
 integration the silex is liable to free solution in water, was con- 
 sistent and satisfactory. But no consistent or satisfactory view 
 could be discovered explanatory of the modes and forms in which 
 the flint is actually found. I endeavoured in the above-cited 
 papers to show that those modes and forms are owing to the very 
 rapid solidification of the siliceous fluid induced by special cir- 
 cumstances combined with the activity of molecular attraction : 
 that those special circumstances were generally the presence of 
 an organic body which acted as a nucleus, the softer bodies 
 being more operative in this respect than the harder : that some- 
 
 * One of the most pleasing duties of the student of natural history is to 
 acknowledge and reciprocate the assistance received from, and always so 
 ready to be rendered by, other naturalists. I take this opportunity of 
 acknowledging my obligations to Professor Owen for, among other things, 
 affording me the opportunity of fully examining several recent specimens 
 of the highest interest, some of which I shall have occasion particularly to 
 mention hereafter; to Mr. Morris, the well-known author of the ' Catalogue 
 of British Fossils,' for assistance rendered in more ways than it would be 
 easy for me to enumerate; to Mr. Tennant of the Strand, and Mr. Harris 
 of Charing in Kent, who have each placed at my disposal several valuable 
 specimens with the liberal permission to make any sections I desired ; and 
 to Miss Emma Naylor of Wakefield for the donation of many fine specimens 
 of recent British Polypifers collected by herself. To many others my 
 acknowledgements are also due, for the loan of specimens, both fossil and 
 recent, for examination and comparison. 
 
 t Lond. and Ed. Phil. Mag. vol.iii. p. 25. 
 
11 
 
 times mere mechanical action was sufficient to induce solidifica- 
 tion. I further showed* that from these facts certain very re- 
 markable and hitherto unnoticed results had followed ; namely, 
 that when a mass of the siliceous fluid had solidified thus sud- 
 denly round a soft body, a Ventriculite for example, in which the 
 soft parts were existing at the time of its envelopment, it neces- 
 sarily formed a solid mould round the nucleus : that when the 
 soft body decayed, its gases and softer components having escaped 
 through or been absorbed by the surrounding and hardening 
 chalk, a more or less complete hollow was necessarily left in the 
 solid flint in the place formerly filled by the animal body : that 
 the firmer fibre of such bodies remaining after the softer parts 
 had thus passed away would afford and did afford centres of 
 crystallization for any silex slowly permeating the stratum in a 
 liquid or gaseous form after that stratum had acquired consistency: 
 that according to the greater or less quantity of such permeating 
 silex would be the result in either wholly or only partially filling 
 up the hollows whose origin I thus explained. I showed the ac- 
 cordance of these views with the known laws of the development 
 of crystals f. 
 
 The most important results follow from these observations ; 
 and I immediately saw that that which had been generally de- 
 scribed as the remains of the body itself was in truth the incrus- 
 tation of a crystallized foreign substance round such remains. It 
 will be obvious how erroneous must be all descriptions founded 
 upon the former notion. 
 
 It remained to adopt a mode of examination of the Ventricu- 
 lidse in chalk and in flint which should realize the living animal in 
 the same form from an inspection of its remains in either matrix. 
 To effect this I made many sections of chalk specimens in every 
 direction, for examination both as opake and transparent objects, 
 an attempt which I believe was novel. Aware that it is by 
 
 * Ann. and Mag. Nat. Hist. vol. xix. p. 306. 
 
 f The mode of crystallization most commonly exhibited in these cases is 
 the acicular. Silex is known to crystallize in this way from sublimation. In 
 the 2nd vol. of the 'Transactions of the Geol. Soc.' p. 274, is a paper by 
 Dr. MacCulloch on an instance observed by him of the actual formation of 
 such crystals, which he there describes aptly as " filamentous crystals cross- 
 ing each other in all directions." Jt seems to me, from careful observa- 
 tion, that it was in cases where an almost or altogether entire hollow was 
 left, all the fibre having decayed as well as the soft parts, that the geometric 
 crystals were more disposed to form within the flints. The centres were 
 fewer, the space larger. A botryoidal form is frequently assumed under such 
 circumstances by the aggregation of acicular crystals ; and traces of its former 
 existence are visible in many now solidly filled spaces. The greater or less 
 degree of slowness of the deposit no doubt had an important influence in 
 determining the silex to assume the geometric form of so-called quartz or 
 the botryoidal form of so-called calcedony ; the quartz being the result of a 
 still slower deposit than the calcedony. 
 
12 
 
 careful personal manipulation only that results can be obtained 
 which may be relied on, I further procured the necessary ma- 
 chinery and made a large number of sections, in every possible 
 direction, of numerous entire flints*. The results have far sur- 
 passed my most sanguine expectations. I have not only dis- 
 covered, beyond possibility of question, the intimate structure of 
 the Ventriculidas, and in so doing have discovered an entirely 
 new and most remarkable form of animal tissue, but I shall be 
 able to show the cause and character of all the modifications of 
 form under which the Ventriculidse are found ; and I further hope 
 to afford indications (I wish to express myself here as cautiously 
 as possible) of the natural affinities and habits of the living bodies 
 to which these very interesting fossils owe their origin. 
 
 It should be remarked, that the difficulties in the way of obser- 
 vation of structure in chalk specimens is no less, in reality, than 
 of those in flint. The very friable nature of the chalk, coupled 
 with the almost invariable presence of oxide of iron, would be 
 sufficient to obscure and practically to obliterate a structure far 
 less delicate than that of these bodies. Hence the coarser su- 
 perficial characters which have been seized upon by all authors 
 as characteristic of genus are all that is usually visible. It was 
 not without much difficulty and care that I succeeded in exami- 
 ning satisfactorily the intimate structure of these bodies as ac- 
 tually preserved in that matrix. 
 
 It will clearly be only by thus gaining an insight into the true 
 comparative values, if I may so speak, of the facts exhibited by 
 remains preserved in both chalk and flint, that the inquirer can 
 be in a position usefully to commence his researches with the 
 hope of reaching any definite and satisfactory results. 
 
 We find specimens of the Ventriculida? preserved in flint in 
 one of three ways : the place of the body either wholly solid, the 
 crystallized silex having entirely filled up the original hollow : 
 partially solid, and that is generally towards the central parts of 
 the flint, the marginal parts, and especially the roots, remaining 
 the last to be solidly silicified : or, lastly, the Whole occupied by 
 an open network only. Each case may be examined with great 
 instruction, the key to what it teaches having first become 
 thoroughly understood upon the principles above indicated, to the 
 deduction of which principles instances like these were steps, and 
 of which they do but afford illustrations. 
 
 It follows from those principles that, where the calcedonyf, 
 whether solid or open, began by crystallization round some re- 
 
 * The sections of flints usually examined by microscopists are of mere- 
 chips and fragments. They can be of no value in an investigation of this 
 nature. 
 
 f See note Ann. and Mag. for May last, p. 307. 
 
13 
 
 maining fibres of the body as centres, the fibre itself, when by 
 this process hermetically sealed up, would remain there to this 
 day in much the same condition as it then was, usually more 
 or less charged with sulphuret of iron ; and that, where not so 
 hermetically sealed, it would, on its subsequent decay, either 
 leave the incrusting calcedony in the condition of a series of hol- 
 low tubes, or such hollow tubes would, still subsequently, be again 
 filled up by a continuation of the process of siliceous crystalliza- 
 tion. I have many specimens of the Ventriculidse in each of these 
 conditions. It will further be obvious, that where, as occasion- 
 ally might happen, but comparatively rarely, because dead fibre 
 would have less attraction for the siliceous fluid than soft animal 
 matter, the fibrous skeleton of an animal of which the soft parts 
 had already decayed away was enveloped in the siliceous fluid, no 
 hollow would ever be formed, but that fibrous skeleton would be 
 preserved hermetically sealed. I have instances of this latter 
 mode of preservation also. 
 
 The flint specimens are, in the vast majority of cases, found 
 with flint on both sides ; a fact resulting, there can be little doubt, 
 from the operation of some cause connected with electricity, 
 which, though there is in no part any communication between 
 the two surfaces, determined an attraction and affinity between 
 equivalent masses of siliceous fluid on the two opposite sides, just 
 as we see the needle follow the magnet though a solid plate in- 
 tervene. Cases are however not very uncommon in which the 
 flint exists in a mass only on one side, there being on the other 
 merely narrow filamentous threads of flint. When the mass is 
 thus found on one side only, that side is, in almost all cases, 
 the inside. Now the Ventriculites being funnel-shaped, there 
 would necessarily be a much greater attraction for the siliceous 
 fluid on the inner side of the body than on the outer side, in pre- 
 cisely the proportionate difference that there is between a single 
 exposed plane surface and a surface closed in on all sides by op- 
 posite surfaces. If a mass of siliceous fluid were not great in a 
 particular spot, the greater part of it would thus be drawn to the 
 inside, the electric attraction before hinted at operating however 
 generally to attract a small portion to the outer corresponding 
 surface, where it would spread in what now appear as filamentous 
 threads between the foldings of the outer surface of the animal, 
 in which spots the greatest attraction on the outside would of 
 course be, in consequence of the opposite surfaces there present*. 
 Occasionally a similar appearance is found both on the external 
 and internal surface ; but that is usually towards the margin of 
 a flint which is otherwise encased on both sides, as in the speci- 
 men fig. 2- of my paper of January, and betokens an exhaustion 
 
 * See and cf. p. 9 and note to p. 301 of the two articles before referred to. 
 The threads of flint are there found though the body was not encased. 
 
14 
 
 of material. It would sometimes happen that a very small mass 
 of the fluid would approach the outside only of a large Ventri- 
 culite. In such case none could reach the inside, and we should 
 find, and do, that it is external only. 
 
 It may at first sight be thought that there would be hollows 
 in chalk specimens similar to those in flint, and that these would 
 be as subject as the latter to have any inclosed fibre incrusted 
 with calcedony. A little reflection will however satisfy the careful 
 reader that this could not be the case. The mould of the indi- 
 viduals inclosed in flint was perfectly solid. Though sulphuretted 
 hydrogen were evolved, which it necessarily would be, from the 
 decomposing body, the encasement in that solid matrix would 
 generally prevent the so free deposit of sulphuret of iron as would 
 take place in the open chalk. Consequently the remaining fibre 
 would offer more affinity for the gaseous silex than would such 
 fibre remaining in the open chalk and more highly charged with 
 sulphuret of iron. And the fact is, that specimens preserved in 
 the chalk exhibit a much larger proportion of sulphuret of iron 
 than those preserved in flint, frequently so much that the spe- 
 cimen, after a short exposure to the air, becomes a mere mass of 
 oxide of iron, and all structure is un distinguishable. In other 
 cases it is less so, and I have in fact found, in several chalk spe- 
 cimens in which the presence of iron is the least marked, that the 
 deposit of crystallized calcedony* pure and without the presence 
 of a particle of flint on either external or internal surface has 
 taken place to a small extent, but in a most exquisite manner. 
 It is clear however that, in general, when any part of the soft 
 substance of a body encased in chalk decomposed, its place was 
 soon filled up with particles of chalk, which in its then scarcely 
 hardened state were readily carried in. The firmer fibre would, 
 thus inclosed, endure for a much longer period, probably indeed 
 until the chalk had become comparatively dry, on which event 
 its absorption, where little sulphuret of iron was present, would 
 follow and the space be left vacant. The instances of calcedonized 
 fibre in chalk specimens are then quite as frequent, and to just 
 such an amount in individual cases, as might be anticipated, while 
 the frequent excellent preservation of the forms of the Ventri- 
 culidse in the chalk is also explained. 
 
 A piece of dead fibrous skeleton in the chalk would evolve but 
 little sulphuretted hydrogen, consequently would induce the de- 
 posit of little sulphuret of iron. It follows that, when buried in 
 the chalk mud, it would be at once closely encased. The hard- 
 ened chalk would, on the absorption of all the components of the 
 fibre, also leave the places of those fibres vacant. On cutting open 
 these, as well as on cutting open specimens encased with their 
 soft parts and whose phenomena it is above attempted to ex- 
 * Not flint : it is important to remember this. 
 
15 
 
 plain, we should find the places of the fibres indicated by hollow 
 tubes disposed through the mass. Of course the presence of 
 great pressure, or the percolation of water or other causes, will 
 have frequently destroyed every trace of the animal, or partially 
 destroyed it, leaving perhaps a mere iron-mould to mark its 
 place ; if the fibre were dead, not even that. I am fortunate in 
 having some very beautiful specimens of the dead fibre tubes in 
 my possession, specimens which naturally escape ordinary at- 
 tention, there being no colour to attract the eye, and the aid of 
 the microscope being necessary to detect the facts. 
 
 In my notes I find several other points examined, but the 
 space to which I am necessarily restricted prevents my entering 
 into further detail here, and I think that I may rely on the can- 
 dour of every competent observer, that, having thus far touched 
 on material points, others which may occur to him have not 
 escaped my attention. Proceed we now to the results of these 
 observations. 
 
 I propose to point out, first, some of the general and most im- 
 portant characters connected with the external form of the body 
 of the living animal ; second, the same of the roots ; third, to 
 show what is the intimate structure of both ; and lastly, to en- 
 deavour to indicate the natural affinities of the whole group. 
 And though it is impossible for me to do otherwise than pain- 
 fully feel that the attempt is vain to convey, by a few words and 
 figures, that certainty of conclusion which I have derived from 
 such very extended observation, I will hope to impart some con- 
 sciousness of a reality. 
 
 Every reader familiar with the human brain is aware that it 
 consists of a very extensive surface folded up in numerous con- 
 volutions in order to pack it in the small compass of the skull, 
 just as for convenience the pocket-handkerchief is doubled up 
 to put it in the pocket. The re- 
 markable resemblance between the 
 annexed section of a Ventriculite 
 (in flint) in my possession, and 
 any cross section of the cerebellum 
 cannot fail to strike every reader, 
 a resemblance arising from the 
 simple circumstance that both 
 are examples of a similar mode 
 
 adopted by nature for packing an extensive surface in a small 
 space*. But no one will pretend to assert that those ridges, of 
 which the outline is seen in the section of the cerebellum, are 
 " cylindrical fibres ; " nor will any one infer a power of expansion 
 
 * Plates 63 B, 64, 64 A, and 64 B of Prof. Owen's Odontography ' afford 
 striking illustrations of the application of the same contrivance to the hard- 
 est, as in the brain it is applied to one of the softest, of organic tissues. 
 
16 
 
 and contraction to reside in these bundles, as Parkinson* and 
 Dr. Mantell have done, from a similar appearance, in respect to 
 the Ventriculida3. 
 
 A glance at the following figure will satisfy any one that a 
 simple and plain membrane may be folded up in the most intri- 
 
 cate way without in fact destroying its simplicity, and having the 
 only effect of packing a larger surface into a smaller compass. 
 The mode of folding may be either longitudinal, in which case 
 we shall have the " cylindrical fibres radiating from the basis to 
 the circumference " of Dr. Mantell ; 'or it may be transverse ; or 
 it may be more or less intermediate between the two, thus 
 causing those longitudinal ridges to appear to anastomose ; or it 
 may be so regularly intermediate as to give to the surface a 
 mammillated appearance. The appearance of the body may even 
 differ on the external and internal surfaces, inasmuch as the folds 
 may assume a different direction as they reach the respective 
 surfaces, as we see familiarly in a rhubarb leaf just burst from its 
 sheath. But none of this can alter the nature of the membrane 
 itself, or serve to establish a generic character. It is further 
 evident that if the folding assume the mammillated character, 
 that is, if a membrane of some thickness be folded in and out in 
 regular figures, a section across any part will, according as its 
 direction shall be, represent a series of apparent tubules or of 
 reticulations regularly disposed, as seen in PI. VII. figs. 1 to 4 ; 
 and again, that if any foreign substance fill up any of the super- 
 ficial depressions on this mammillated surface, that surface will 
 appear to be regularly perforated by tubules. Hence the figures 
 of Goldfuss and Roemer, which profess to give magnified views 
 of the exteriors of Ventriculidse. The following figure will illus- 
 trate this : a, a, is the membrane, which is folded in and out with 
 
 Fi. 
 
 &L 
 
 the exactest regularity, each fold being of equal breadth and 
 * Organic Remains, ii. p. 145. 
 
17 
 
 length, thus making each depression and elevation regularly 
 round. It is evident that a section made in any part from b to 
 by in the plane of the surface, would be more or less of the cha- 
 racter of PL VII. fig. 3, while a section made in any part perpen- 
 dicular to that plane would resemble the upper part of fig. 4, or 
 fig. C (p. 15). 
 
 A very careful examination and comparison of innumerable 
 entire specimens and sections, both in chalk and flint, of eveiy 
 form and variety of Ventriculidse, early satisfied me that, though 
 contrary to every figure and description that has been published, 
 the above was the true explanation of their forms. Every sub- 
 sequent observation has confirmed this conclusion. Let the 
 reader examine PI. VII. figs. 1 and 2, and fig. C (p. 15), and he 
 will see three out of many of the modes in which transverse sec- 
 tions of Ventriculidse -in my possession show that the membrane 
 is folded. Let him, again, examine PL VII. figs. 1 and 2, and he 
 will see in fig. 1 a surface perfectly plain and smooth, except so far 
 as the fibrous structure marks it ; while in figure 2 he will see the 
 lower part of the surface smooth like fig. 1, but the upper part 
 gradually passing into the mammillated character. These varie- 
 ties (never before figured) are both in my possession, and I have 
 every shade and variety from the perfectly plain to the most con- 
 voluted form of fig. A above. It will easily be conceived what 
 a variety of markings in the chalk and flint, at first sight wholly 
 inexplicable and unconnected, will be presented on longitudinal 
 and transverse sections of such complicated convolutions. 
 
 Nor does such variation in external form indicate in the least 
 degree a habit or power of contractility. I have shown, by 
 analogy, that it does not necessarily do so. I will show by ac- 
 tual facts that the Ventriculida3 were not contractile. Not only 
 is it an important fact that we find deeply convoluted specimens 
 as wide-open-spreading as any, while we find specimens of the 
 perfectly plain varieties as narrowly funnel-shaped as any, but we 
 find, fixed on the surfaces of specimens of every variety, Ostrese, 
 Dianchorse and other shells which are in the habit of attaching 
 themselves by peculiar processes to other bodies. In every such 
 case the shape of the attached shell departs from its ordinary form, 
 and is precisely adapted to, moulded on, that of the Ventriculite 
 proving that it grew thereon, and thus testifying at once to the 
 firmness of the texture of the body and to its noncontractility, as 
 well as to its durability. Further, we sometimes find delicate 
 Flustrse spread over parts of Ventriculites always, as before, 
 without trace of distortion or disturbance. But there is, if pos- 
 sible, yet stronger evidence in the fact that an entire and very 
 large group of the Ventriculidse, and those, too, the very ones 
 which are pointed out as exhibiting the most " contracted state/' 
 
 B 
 
18 
 
 have distinct heads which heads are perfectly smooth and re- 
 gular in general form, and with no deep anfractuosities, instead of 
 being, as the contractile theory requires, most of all convoluted. 
 Finally, I shall presently show that there was a special and most 
 beautiful provision in the intimate structure of these animals 
 against contractility either voluntary or by ordinary accident. 
 
 Nor is there any want of examples of the presence of similar 
 convolutions without contractility in what I hope to show to be 
 kindred zoophytes of the present day. The Eschara foliacea is 
 familiar to every one. But I am indebted to the kindness of 
 Professor Owen for a still more striking illustration of this ar- 
 rangement in a recent zoophyte. That gentleman a short time 
 ago placed in my hands, with the liberal permission to examine 
 it in any mode I chose, a specimen of Meandrina recently brought 
 from the Indian Ocean, and which had been treated with mu- 
 riatic acid. All the calcareous parts being thus dissolved, there 
 remained only the soft animal parts. I observed with much 
 gratification that these consisted, in fact, of a single membrane 
 folded up almost exactly after the manner of some of the more 
 irregular of the Ventriculites very much indeed like the Ventri- 
 culites radiatus described by Mantell and others as having an 
 " integument formed of cylindrical fibres anastomosing," &c. 
 
 However much differing in the complexity and mode of the 
 convolution of its membrane, the body of every member of the 
 family of Ventriculidse appears to have had an opening at its 
 upper part, and that body approached more or less to the form of 
 an inverted cone. They usually grew single. In one or two 
 species they are grouped, and there are occasional instances, but 
 very rare, of double specimens of those whose usual habit is single 
 just as we occasionally find a double Actinia. 
 
 Having thus shown the general character and habit of the body 
 of these animals, I next proceed to show the nature of the root. 
 
 The attention of Ellis was particularly attracted, in describing 
 his Corallines, to one variety, the Corallina astaci corniculorum 
 amulet as he calls it, or Lobster's Horn coralline, as having roots 
 very different from those of ordinary corallines, ' ' which rise up 
 from an irregular mass matted together to form the stem *." The 
 roots of this zoophyte, on the contrary, tf regularly enter in in 
 whirls round the joints ;" the body of the animal, according to his 
 figure, tapering off to a point in the midst of these root fibres f. 
 
 Something after such a type was the habit of the roots of the 
 Ventriculidse. They were not, as described by Dr. Mantell, de- 
 
 * Ellis's Corallines, p. 16 and pi. 9. B. 
 
 f- This peculiarity is not marked either in the description (p. 86, 2nd ed.) 
 or figure (pi. 19) of Dr. Johnston; but the figures and descriptions of Ellis 
 are seldom unworthy of dependence. 
 
19 
 
 rived from the base of the body " prolonged into a stirps and 
 attached to other bodies." The body of the Ventriculite tapers 
 off regularly to a point at the bottom. At about an inch from 
 its base, the distance varies according to size, fibres of a very 
 different aspect from those composing the membrane of the body 
 begin to be attached to the external part of that body. They do 
 not begin all round at exactly the same distance from the base, 
 nor, as they increase in mass, is that mass of the same thickness 
 on all sides. They are at first very few and thinly disposed. As 
 the base narrows they increase rapidly in number and mass, till, 
 immediately below the base, they form a bundle of considerable 
 size, which is continued, thus united, for some distance, from one 
 to three inches according to circumstances, when it divides into 
 several radicles, sometimes more, sometimes less : I have counted 
 as many as forty in one specimen. These sometimes extend very 
 far and always terminate in very delicate extremities. 
 
 The root-mass is never itself convoluted like the body, even in 
 its upper parts, and where it forms a thin membrane only. It 
 necessarily follows the form of any convolution, usually slight, 
 which exists at the lower part of the body. The mass of the root 
 is not regularly cylindrical, but irregular on the different sides 
 (see PL VII. fig. 6). Occasionally it assumes a tubular aspect in 
 places, but this is quite accidental and in nowise characteristic 
 of genus or species. 
 
 PL VII. fig. 5 gives a longitudinal section in which the differ- 
 ence between the body and root is very clearly seen : fig. 6 is a 
 transverse section of the same specimen. These two will realize 
 to the observer how admirably the body was lodged in this root- 
 case as in a sheath or socket. They remind one of a balloon to 
 which the car is attached by a network of rope gathered in on all 
 sides round the narrow base. PL VII. fig. 7 is the external appear- 
 ance of another specimen, showing the root-fibres commencing 
 round the body and spreading at the root. 
 
 The substance of the root was different, as the fibres were dif- 
 ferently arranged from those of the body. This is proved by 
 the fact that the root exists generally in a very different condi- 
 tion, both in chalk and flint, to that in which the body is found 
 in the same specimen. In flint its place is much less often 
 solidly filled with calcedony than is that of the body. In the 
 chalk it is the rare exception to find the substance of the root in 
 sound condition. It generally falls away to dust the moment 
 the specimen is opened, while the body of the Ventriculite re- 
 mains entire and perfect down to its very point, having all around 
 it a hollow space where the root was. 
 
 The Ventriculites were never " affixed to other bodies." It is 
 possible that the animal was locomotive like the Actinia, and that, 
 
 u2 
 
20 
 
 like thePennatulida, it fixed itself during pleasure in the soft mud. 
 Among all the thousands of specimens which I have examined, I 
 have never seen one attached to a shell or to any other solid 
 body. Shells are indeed sometimes found growing onto the upper 
 part of the root, where it was of course immoveable, as they do 
 on the body. The most delicate terminations of the roots m ay 
 be always traced by their impression in the chalk. If the ani- 
 mals were locomotive, it was by aid of the lower radicles that 
 they progressed. 
 
 Occasionally, but it is the rare exception, a small bundle of 
 root fibres is given off from the side of the animal. This is 
 similar in character to the true root, but slighter, and it is 
 merely affixed by apposition of its thick extremity, without any 
 of the encasement. Such instances appear as if some circum- 
 stance rendered an additional support necessary in the particular 
 instance. There are instances among our native zoophytes of 
 upright habit, in which a similar circumstance is not uncommon 
 much more common than among the Ventriculidse. 
 
 I have one example in which there is no encasing root-sheath. 
 And in that specimen, as if to make up for want of it, the root- 
 lets begin to arise about an inch from the base, being already of 
 considerable strength : they spread out immediately on each side, 
 and so support the body just as a tent is supported by the stay- 
 ing ropes on all sides. I have another example having two com- 
 plete roots, and of course a divided base to the body. Each root 
 is however smaller than usual. These examples only show that 
 the world was under the same laws in the days of the chalk for- 
 mation as it now is : that then, as now, monstrosities would some- 
 times appear, which, however, only themselves serve to show the 
 permanence of Law and Unity, inasmuch as in these very mon- 
 strosities there is always present some compensating pheno- 
 menon which it is interesting and instructive to observe. 
 
 I have already stated that my investigations into the inti- 
 mate structure of the Ventriculidse have rewarded me by the dis- 
 covery of an entirely new kind of animal structure. 
 
 In 1841 Professor Owen read before the Zoological Society a 
 paper* on a remarkable production from the Philippine Islands, 
 of which he says, " in the exquisite beauty and regularity of the 
 texture of the walls of the Cone, the species surpasses any of the 
 allied productions that I have, as yet, seen or found described," 
 and he gives to it the very appropriate name of Euplectella. 
 While the Euplectella is the only object which approaches the 
 Ventriculidse in the beauty and regularity of its texture, the 
 latter far surpass it not only in possessing a much higher degree 
 
 * Zoological Transactions, vol. iii. p. 203 ; and Ann. and Mag. Nat. Hist, 
 vol. viii. p. 222. 
 
21 
 
 of that very beauty and regularity, but in the exquisite delicacy 
 of a further texture, of which the Euplectella does not possess a 
 trace, and which, so far as I can learn from the best authorities, 
 has never been hitherto observed in any object, animal or vege- 
 table, recent or fossil. 
 
 Through the kindness of Professor Owen and the liberality of 
 the proprietor of the specimen, Mr. Broderip, I have had an op- 
 portunity of carefully examining the Euplectella. While I am 
 thus enabled to speak to the fidelity of Prof. Owen's description, 
 I can also speak with more assured certainty as to points to 
 which, having the structure of the Ventriculidse before my eye, 
 my attention was more particularly directed. 
 
 The Euplectella is composed of an arrangement of fasciculi of 
 fibres, one course over the other, not anastomosing together : its 
 principal substance consists of two layers, a longitudinal and a 
 transverse one, of which the former is external to the latter. 
 These, thus crossing at regular intervals, form the regular tex- 
 ture which excites Prof. Owen's just admiration, an arrange- 
 ment in which they are held by the interlacement of other and 
 more delicate fibres tied by the same fairy fingers which Dr. John- 
 ston describes as knitting the plexus of one of his delicate zoo- 
 phytes. The squares thus formed are about the eighth of an inch 
 in size, gigantic in comparison with the squares filling the 
 membrane of the Ventriculidse. 
 
 The membrane of the Ventriculidse is composed of very de- 
 licate fibres arranged in squares, of which the larger ones measure, 
 on an average, considerably less than the 100th of an inch on each 
 of their sides *. The fibres of the Ventriculidse are not arranged in 
 fasciculi, nor in layers the one overlying the other concentrically or 
 otherwise. Neither are they at their points of crossing wrapped 
 together by other interlacing fibres. The substance of the Ventri- 
 cuiites simplex (see PL VIII. fig. 1), which is the true type of the 
 whole family, is not quite one line in thickness (about the sixteenth 
 of an inch) . It consists of five thicknesses of the squares 1 have 
 mentioned. Consequently the substance of the body of the ani- 
 mal consists of a membrane composed of an exceedingly delicate 
 fibre anastomosing in every direction, so as to form both in the 
 plane of its surface and of its thickness regular squares (see 
 Plate VII. fig. 8) . As the body increases in size from the base to 
 the upper margin the fibres increase in number, and this takes 
 place, not, as in the Euplectella, by the " convergence and inter- 
 blending of contiguous fibres," but by the addition of a fresh fibre, 
 generally at the middle of the outward boundary fibre of a 
 
 * The sizes are pretty constant in different species. There is some va- 
 riation however. Those described are the largest. The relative dimensions 
 appear constant. 
 
22 
 
 square, thus causing less disturbance to the regularity of the 
 squares than happens in the Euplectella (see fig. 14). But this 
 is nor all the beauty and delicacy of the intimate structure of 
 the Ventriculidse. A structure remains to be described, to which 
 the expressions of Prof. Owen in describing the Euplectella may 
 well justify me in saying that no language can do justice, and 
 which no one can contemplate without delight, wonder, and 
 exquisite admiration. 
 
 If the reader's attention has ever been attracted by the roofs 
 of the large railway stations, he will have perceived that they are 
 held together by the mutually counteracting and balancing effects 
 of thin rods obliquely placed any one of which would singly 
 be very inefficient for any substantial purpose. To give a barred 
 gate strength, or to keep a loose door or window-frame to its 
 true square, we see the carpenter fix a bar obliquely subtending 
 the right angle, which will hold the more securely the nearer it 
 is fixed to each side at equal distances from the angle. The prin- 
 ciple of the bracket which supports a shelf or bust is but the same. 
 
 But there is nothing new under the sun. Ages before rail- 
 ways or carpenters existed, nature had adopted this very plan, to 
 give strength and stability to the deep ocean forms of the whole 
 family of Ventriculidse ; only, as she ever does, adopting a method 
 far more delicate, complete and beautiful than it were possible for 
 the hand of man to execute. 
 
 I have said that the fibre is arranged in a tissue of regular 
 squares, which are formed by the anastomosing, at each angle, 
 of that fibre. But, besides these fibres, of which there necessa- 
 rily occur at each angle three entire ones crossing each other, or 
 six looking at them as radiating from the angle as a centre, these 
 crossing and anastomosing fibres are strengthened and secured by 
 twelve still finer oblique fibres, each about one-fourth of the length 
 of any one square itself. Each of these fibres subtends one of the 
 angles formed by two of the primary crossing fibres, and of which 
 angles there are of course at each crossing twelve. Each of them 
 anastomoses with one pair of the primary fibres in each position 
 in which they meet to form a right angle. This anastomosis 
 takes place at an equal distance on each primary fibre from the 
 angle itself, namely, at a distance of about one-fifth of the length 
 of one side of the square. Thus it will be seen that at each 
 place of crossing there are twelve subtending fibres and six pri- 
 mary fibres, in all eighteen fibres. Now, taking a Ventriculite 
 of very moderate size, say three inches in height and plain, we 
 shall have a membrane containing at least 750,000 squares, and 
 at least nine million of these delicate subtending fibres, each 
 faultless. What a marvellous piece of workmanship is this ! 
 
 It will be perceived that by this most admirable contrivance a 
 
23 
 
 regular octahedron is formed round about the point of union of 
 every four squares throughout the whole body of the Ventri- 
 culite, thus forming, of course, as many octahedrons as there 
 are squares. See Plate VII. fig. 8, and more clearly fig. 10, as 
 seen in flint, and fig. 9 as seen in chalk*. 
 
 It is needless to point out the strength given to the whole 
 membrane by this arrangement, and the obvious design of it to 
 prevent any injury of the animals to which it belonged arising 
 from any yielding or distortion to which it might otherwise be 
 liable from the operation of ordinary causes present in the ocean 
 where they dwelt. They generally suffered fracture rather than 
 yield to such impulse. 
 
 As the texture reaches the surface, both external and internal, 
 it assumes a different appearance in order to attach to it the 
 more securely what I shall crave permission to call, for the pre- 
 sent, the polyp-skin. The regular squares and their octahedral 
 junctions are still present, but under a different aspect ; their size 
 being contracted to about the 300th of an inch. This appears 
 to be effected by the addition of finer fibres crossing each other 
 within each square, so as to make at least four squares equivalent 
 to each of the larger ones. This membrane presents a solid 
 series of these squares ; that is, they extend, in a single layer, as 
 well in the plane of its thickness as of its superficies. The 
 membrane spreads over every convolution and descends every 
 anfractuosity of the body. Plate VII. fig. 11 shows this membrane 
 in calcedony, where several of the actual fibres are preserved and 
 the crystallization round them is very visible. 
 
 External to the whole, the polyp-skin itself, spread over 
 both external and internal surface and depressed also into each 
 of the anfractuosities, stretches over the delicate membrane last 
 described, with which it is closely united (see Plate VII. figs. 12 
 and 13, and PL VIII. fig. 6) . In specimens both in flint and chalk, 
 prepared with care and in a high state of preservation, an equi- 
 distant row of apparent denticles seems to extend from the inner 
 substance to the external encasing wall. This is caused by the 
 transverse fibres (in the flint, incrusted with calcedony ; in the 
 
 * It is obviously impossible to find individual specimens which shall 
 fully show those states of fact which have been only ascertained by careful 
 examination of many hundred specimens. Again, the least obliquity in 
 any section will cause an apparent elongation of some and a cutting oft' of 
 parts of other squares ; while the slightest variation in focus will cause fibres 
 in different planes to appear on the same plane. Hence, oftentimes, an ap- 
 parent irregularity which does not really exist, but which may easily deceive 
 an unpractised eye either in the object itself or in the engraving. Figs. 8, 1 1, 
 13 and 14 illustrate this. It should be added, that the octahedral structure 
 could not be given with any clearness in figs. 8, 9 and 14, without some- 
 what exaggerating its relative size. No engraving can do justice to the 
 exquisite delicacy of the original. 
 
24 
 
 chalk, the place of the fibre left hollow and the intervals filled with 
 chalk) of the delicate membrane which underlies the polyp-skin. 
 
 The polyp-skin itself is of extreme tenuity, and differs alto- 
 gether in structure from the internal parts. It is not fibrous, but 
 of a uniformly close texture which yields to the highest powers 
 of the microscope no other than a minutely granular appearance. 
 
 It is only in very rare cases that the polyp-skin is found in 
 any degree of perfection. Careful observation and comparison 
 of an immense number of specimens, with their casts, led me to 
 infer the existence of this polyp-skin, as matter of unevadable con- 
 clusion, long* before I was fortunate enough to find, as I have 
 since done, specimens in which it was so preserved as to be found 
 on the body of the fossil itself. In such cases as the latter, the 
 internal structure is always clearly seen where any parts of the 
 polyp-skin are ruptured. See PI. VII. fig. 12. 
 
 I shall presently enter on the consideration of the polyps 
 themselves and the accompanying phenomena. 
 
 It will of course be understood that the membranes I have de- 
 scribed must, during the life of the animal, have been filled with 
 soft parts, which, with whatever minutely ramifying vessels they 
 contained, rapidly decomposed, and of which therefore I have as 
 yet discovered no trace. 
 
 There existed no spiculse, siliceous or calcareous, in any of the 
 Ventriculidae ; some spiculse in the adjoining chalk or flint may 
 sometimes deceive an inexperienced observer. 
 
 The structure of the root differs much from that of the body. 
 Annexed to the body by short fibres clearly seen on a good sec- 
 tion, it is, like the body, arranged in squares, but those squares 
 have not all that regularity which those of the body have. They 
 are, in general, smaller in their average size than those of the 
 body, and have throughout a tendency to elongation in their 
 longitudinal direction, whence they are often narrowed on one 
 or both of their lateral planes. The fibres of the root are also 
 thicker in their longitudinal direction than in their transverse. 
 
 It is particularly important to observe that there is no trace 
 in the roots of the octahedral structure, a fact precisely in ac- 
 cordance with the functions of those roots. The safety of the 
 animal would be more secured by the latter yielding to every 
 impulse, and waving their so delicate load from side to side, than 
 by an unbending stiffness. The octahedral structure had there- 
 fore no place in the roots. 
 
 The integument of the root was also very different from that of 
 the body. It is impossible for a moment to confound the two. Its 
 longitudinal fibres were much thicker than the transverse ones ; 
 and, there being no octahedral structure to secure an unyielding 
 exactness, it appears as if disposed in long, narrow, and not very 
 regular meshes. When encrusted with calcedony it appears not 
 
25 
 
 unlike some vegetable tissues (see fig. 7. PL VIII.). The decep- 
 tiveness of appearances thus caused has been already fully 
 pointed out. This integument was very possibly muscular. 
 
 The fibres of the roots, like those of the body, all anastomose 
 together. They do not overlie nor entwine. 
 
 When it is remembered that sulphuret of iron is deposited 
 more or less on every fibre that has been actually preserved, it 
 will be obvious that it is impossible to ascertain the exact size of 
 the recent fibre. I have however frequently observed the better- 
 preserved fibre to be less than the 4000th of an inch in diameter 
 in its present condition ; much appears about the 2000th of an 
 inch ; and the coarser and less perfectly preserved rarely exceeds 
 the thousandth of an inch in its present condition. The fibre is 
 single and solid (never fistular). It is generally found both in 
 flint and chalk reduced, more or less, to its ultimate granular tex- 
 ture, in which case it resembles the granular texture of other 
 animal fibre. This granular texture is finer, even in its present 
 condition, than that of the recent Actinia. 
 
 The description of another most interesting point which I have 
 discovered in connexion with these animals the ovarian cells 
 will more properly come under the next division of the subject, 
 when the natural affinities of the animal are considered. I con- 
 tent myself for the present with stating the fact of the discovery 
 and clear establishment of these ovarian cells, a fact which cannot 
 but be felt by every naturalist to be of the very highest import- 
 ance, both in relation to the individual beings themselves and as 
 an aid in determining their natural affinities (see PL VIII. fig. 3). 
 
 And now, having thus too imperfectly described the intimate 
 structure of these animals, so elegant and graceful even in their 
 external forms, I hope that I shall be felt not to have expressed 
 a too strong sense of the exquisite beauty of that structure. I 
 have searched in vain amid zoophytic forms for any structure 
 that may compare with that of the Ventriculidse in delicacy, 
 beauty, and obvious adaptation. The pride of man may call all 
 those beings who differ most from him in structure the " lower 
 animals ;" but I would ask, Where can be found an organization 
 more complex or more exquisitely delicate, or where adaptation 
 more perfect, than is displayed in the structure of the Ventri- 
 culidse ? Where can we find a structure affording more conclu- 
 sive evidence of the all-prevalence of those laws of Unity and 
 Design which it is the grateful task of the naturalist to de- 
 velope, and of which his inquiries, the further they extend, do 
 but unfold a wider field of illustrations for man to study and 
 admire ? 
 
 The structure which has hitherto engaged our attention is, 
 strictly, that of the polypidom only. It remains to seek the na- 
 
26 
 
 tural affinities of the group the polypidom of which is marked 
 by these high characters of structure. 
 
 The difficulty of this task will be readily appreciated when it is 
 considered that it involves an inquiry into the nature, in respect 
 to fossil bodies, of the parts most perishable and most difficult of 
 observation even in recent species, and that it becomes necessary 
 first to remove preconceived notions, and afterwards to begin de 
 novo to find out the true relations. Further than this, in the 
 structure which I have already described, the combined process 
 of induction and direct observation has enabled us to arrive at 
 conclusions which, after the work has once been carefully done, 
 are definite and positive and incapable of being disputed with any 
 show of reason. The present inquiry must depend more entirely 
 upon the process of induction alone, and I cannot but be con- 
 scious that conclusions thus arrived at, however satisfactory to 
 some, will be always more open to dispute. 
 
 Late writers on zoophytes have been in the habit of treating 
 slightingly a reference to the polypidom in determination of ge- 
 nus and species, and of alleging that little or no reliance can be 
 placed upon anything but observation of the polyps themselves. 
 I apprehend, however, that this originates in some misconception. 
 The Mastodon and the Mylodon are creatures which are well 
 realized to the mind, though no part of the softer tissues of either 
 has been ever seen by their describers. In the ' British Fossil 
 Mammals ' are vivified before us hosts of animals, hardly one of 
 which has ever been touched by the hand or seen by the eye of 
 man in other than its harder parts. How is this ? We owe it 
 to the ever-present and strong-felt conviction of the all-prevalence 
 of the Law of Unity, whence their describer would as soon doubt 
 the evidence of his senses as the important truth that there are 
 always present certain constant relations between certain parts. 
 Now either the Law of Unity and the whole principles upon 
 which the British Fossil Mammals have thus been vivified are 
 mere empty words and baseless, or the same principles must apply 
 to every part of the animal kingdom. It would be as reasonable 
 to deny the value of comparative anatomy as applied to fossil 
 mammals because they who first found mammoth bones imme- 
 diately saw in them glorious evidence of the reality of " those 
 days " in which " there were giants on the earth," as it is to treat 
 as of slight value the forms and structure of the polypidoms of 
 zoophytes because Lamouroux and others may have erred in 
 their application of what they saw or thought they saw in them. 
 The cause of the false deductions (they have not been real in- 
 ductions) drawn from observation of polypidoms appears to me 
 very simple. Keeping now to fossil forms and turning over the 
 pages of Goldfuss and others, no one having any knowledge of 
 
27 
 
 the subject can fail to perceive that, in the hurry to name and 
 figure as many objects as possible, the merest superficial charac- 
 ters of external form have been alone observed, without the least 
 examination of the true characters of those forms or their true 
 texture and habit. I have so fully shown already the necessary 
 consequences of this course in treating of the folding up of the 
 membrane of the Ventriculidse that it is unnecessary to enlarge 
 upon it here. The result has been that, in Goldfuss for example, 
 we have heaped together, without order or method, under the 
 name of Scyphia (for instance), an immense number of objects 
 differing totally from one another in every character, and many 
 of them having no one essential character in common with the 
 definitions of the genus. From the same hurry in jumping to 
 conclusions, and eagerness, at all risks, to name and figure, has re- 
 sulted the disposition in our own country to call everything which 
 it is found difficult to understand and troublesome to investigate 
 either a sponge or a coprolite. We have seen that the Ventri- 
 culidse have, like many other as widely different things, been 
 thrown into the former category, and I have little doubt that, 
 by others, they have been thrown, with equal correctness, into 
 the latter. 
 
 Observation then of mere external superficial fprm, as it is pre- 
 sented in recent specimens, still more as it always must be pre- 
 sented in fossil bodies, whether preserved in a hard or soft matrix, 
 can never lead to other results than we have seen to have been 
 arrived at by the numerous band of writers already named. By 
 such a course we can only again have, as we have had already, a 
 soft-bodied animal gravely described and figured by even such 
 an author as Blainville as having a central stony axis with deep 
 tubular cells the so-called axis being the chalk or flint which 
 had filled the central cavity, and the so-called tubular cells being 
 simply the folds of the membrane forming the polypidom. The 
 same care is necessary to investigate the true structure, character 
 and habit of the polypidom as those of the polyps. Without the 
 nicest manipulation and observation neither can be known or 
 understood. And as without the aid of every artificial means the 
 very existence and nature of the whole class of recent Polyzoa 
 could have been never known, so the polypidoms of fossil zoo- 
 phytes can never be known or understood unless we go to work 
 to vary the words of Van Beneden applied to a kindred sub- 
 ject with the hand upon the slitting-wheel and the eye upon the 
 microscope. It is important too to remember, that even those 
 means are vain unless we have first made ourselves masters of the 
 nature and origin of the matrix itself in which the fossils are pre- 
 served. Without that preliminary, nothing but confusion and 
 contradiction will appear in the observations. Without it, it will 
 
28 
 
 be sufficiently evident that the important and satisfactory results 
 already arrived at as to the structure of the polypidom of the 
 Ventriculidse never could have been attained, and without it they 
 never can be verified. 
 
 I cannot give a more conclusive or satisfactory illustration of 
 what may be done towards attaining a correct knowledge and 
 just classification of the polyps themselves by a careful inves- 
 tigation of the true nature and structure of polypidoms super- 
 ficially very different, than by reference to the labours of Milne 
 Edwards in respect to the Tubulipores*. That writer was enabled, 
 chiefly by this means, to correct the errors of numerous former 
 writers, and to show the true affinities of objects theretofore 
 grouped in most widely different relations. 
 
 If these observations are well-founded, it will follow that at- 
 tention to the nature and structure of polypidoms in general is a 
 matter of much higher importance in determining the true cha- 
 racter and affinities of zoophytes than is generally admitted, and 
 it will appear that a true and careful examination of the combined 
 characters [not any individual points] of structure of polypidoms 
 will afford as safe and important guides in determining the affi- 
 nities, and thereby leading to a just classification, of zoophytes 
 as is afforded by fossil bones to the comparative anatomist in his 
 attempt to vivify the long-departed forms of mammals. No one 
 imbued with the true idea of the Law of Unity, without which 
 science is a mere name, can doubt but that in the one case as in 
 the other, constant relations must exist between one part and 
 another of the entire organization of the recent animal f. 
 
 Nor can this reasoning be in the least degree affected by any 
 theory as to the mode of formation of the polypidom. This will 
 be too obvious to need more than bare allusion. 
 
 The importance of the application of these principles in the 
 investigation of a class of objects like the zoophytes, of which 
 the polyps themselves are so perishable, will be sufficiently evi- 
 dent to the student either of the recent or fossil forms 01 this 
 most interesting family. 
 
 To apply these principles in the present case. In the poly- 
 pidom of the Ventriculida3, to some of the principal and more 
 striking features in the structure of which our attention has pur- 
 
 * Ann. des Sciences Naturelles, April 1838, p. 193. 
 
 f Thus the " affinity " named by Dr. Johnston as existing between the 
 polypidoms of Alcyonlum and /llcyonidium is only " apparent," being of the 
 merest superficial character. Hence it is that I use, below, Dr. Farre's 
 name of Halodactylus instead of that of Alcyomdium ; not because I like the 
 prevailing taste for Greek names, but because the two names Alcyonium and 
 Alcyomdium are so nearly alike as almost necessarily to engender erroneous 
 ideas of an aflinity between the animals to which they have been applied. 
 
29 
 
 posely been hitherto alone directed, many remarkable characters 
 have been found which distinguish it at once from the polypidom 
 of any zoophyte, recent or fossil, hitherto described. At the very 
 first step of our inquiry into affinities we seem then, at first sight, 
 to be put at fault. And this is true in so far as that it becomes 
 obvious that we can range these bodies among those of no recent 
 genus or even family. But that point alone is something defi- 
 nite. And further, the character of a difference is oftentimes a 
 guide where points of analogy fail. In what respect does the 
 structure of the polypidom of the Ventriculidse differ from that 
 of all other known fossil and from all known recent forms ? It 
 differs in the much higher character, the much greater complexity 
 and delicacy of its organization ; in the peculiar adaptation in it 
 to the nature and purposes of polyzoic life. It has been seen that 
 the polypidom of the Ventriculidse exhibits, in the most marked 
 and exquisite manner, structure, in the true sense of that word 
 and as opposed to that mere mechanical arrangement which is ex- 
 hibited by most vegetable and many animal tissues, and which 
 results from the forms taken, more or less regularly according to 
 circumstances, by the mere close compression against each other 
 of circular cells, which causes them to assume a hexagonal, square 
 or irregularly angular form. The octahedral structure of the 
 Ventriculidse on the contrary must take rank with the highest 
 forms of animal structure, as one of the most striking as well as 
 the most elegant and delicate of the evidences which the animal 
 kingdom can afford of design and adaptation. 
 
 The first conclusion then which must be drawn, if the prin- 
 ciples above indicated be sound, is that the being for whose sup- 
 port this so delicate framework was organized must have be- 
 longed to a very high order ; to a higher order than any known 
 zoophyte recent or fossil. 
 
 With an indication of this kind before us, the next legitimate 
 step in the inquiry seems to be to ask, are there any individual 
 characters which distinguish any among the higher orders of 
 zoophytes of which any traces can be found to have been preserved 
 to us in the fossil forms of the Ventriculidse ? 
 
 And by a different process we are brought pretty nearly to the 
 same point. It will be useful, as a confirmation of the value of 
 the test above applied, briefly to allude to this. When it is 
 found that the so-called (at one time) polyp-cells and (at another) 
 absorbents are neither the one nor the other, the inquirer has 
 lost hold of any hint or suggestion from any previous inquirer as 
 to the true character and affinities of these bodies. He asks first, 
 on entering on the inquiry for himself, whether there be no cha- 
 racters in common between these bodies and the single and simple 
 zoophytes of the present day or the more simple of the compo- 
 
30 
 
 site zoophytes. But he finds nothing which can connect them 
 with any of either the Hydrozoa or Anthozoa of Owen (Anthozoa 
 of Johnston). The structure of the polypidom of the Alcyonium, 
 which it is desirable specially to mention as these bodies have 
 been so often referred to that affinity, is found to be essentially 
 different. If he refer to the excellent figure and description 
 given on p. 176 of Johnston's ( Zoophytes*/ or, still more conclu- 
 sively, if he examine the two admirable preparations by Mr. 
 Goadby in the Museum of the Royal College of Surgeons, and 
 which exhibit sections, transverse and longitudinal, of the same 
 Alcyonium, he will perceive an arrangement of canals or tubes 
 bound together by an irregular network, and will at once perceive 
 that in no one point does this resemble the structure of the Ven- 
 triculidse, in which there are no tubes whatever f. He is thus 
 necessarily led to the same point, to which, by another means, we 
 have already arrived, though with less suggestiveness of the true 
 position of the objects before him. 
 
 Starting then from that point, it must be always remembered 
 how much has been already determined as being essential in con- 
 tradistinction to any accidental or specific characters of these 
 objects J. It has been shown that they formed, though soft- 
 bodied as distinguished from stony or from a toughly flexible 
 
 Second ed. 1847. It is proper further to remark, that the characters 
 assigned by Dr. Johnston (p. 174) as generic coincide with this figure and 
 description. It must be presumed that those generic characters have not 
 been assigned without a sufficiently extended observation, not confined alone 
 to British species. 
 
 t I would refer to note, p. 35, for another marked and characteristic 
 difference between the polypidom of the Ventriculidae and any of the 
 Anthozoa. 
 
 | Some, even shrewd observers, appear to find the greatest difficulty in 
 divesting themselves of the influence of ideas arising from mere general 
 external shape. This is extraordinary when there is no one branch of com- 
 parative anatomy, not to say natural history, in which its little importance 
 is not evidenced, and errors arising from it being daily exposed. They 
 point to the cupshaped sponges as so very like the Ventriculidae. I admit 
 a partial superficial likeness to some forms of Ventriculidae, but that is all. 
 And it may be a relief to such objectors to remind them of the external 
 form of several Polyzoa, e. g. Tubulipora patina, which is described by 
 Johnston, p. 267, as having a " polypidom like a little saucer : it varies a 
 good deal in the deepness of its centre, for sometimes it is properly described 
 as being cupped, at other times it is so shallow that a saucer or plate becomes 
 the best object of comparison." There is a real analogy here to the external 
 form and habit of the Ventriculidae : and see the formsof T. truncata, T.peni- 
 cillata, &c., all much more like some Ventriculidae than any sponges are ; and 
 Milne Edwards points out a development of roots (ut ante, p. 211) which 
 forcibly recalls the Ventriculitic root. In the Museum of the College of 
 Surgeons there has been deposited, within the last few weeks, a most 
 superb coral from the Indian Archipelago which precisely resembles in its 
 general external form some of the more usual forms of the Ventrioulido 1 . 
 
31 
 
 sponge-like mass, a firm and enduring body, not contractile*, 
 and with a peculiar structure for maintaining its normal shape. 
 The variety of actual form under which they are found being 
 very great, it must be particularly remembered also that any 
 essential character dependent on any apparent tubules, external 
 or internal, is most distinctly and unequivocally negatived. Were 
 other proofs wanting, this point is established by the fact that 
 such apparent tubules are found in some few special forms only, 
 while in the greater number of the forms of Ventriculidse all 
 marked, in common with those last-mentioned^ by those cha- 
 racters of structure above described, and which therefore must 
 necessarily constitute the essential and generic characters of the 
 whole group all trace of such tubules is absent. Some charac- 
 ters must therefore be found common to all these forms, or 
 common to many different forms, and the apparent absence of 
 which, if undetected in others, can be explained, before we can 
 hope to have got any hint pointing to the affinity of these bodies 
 with any recent forms. 
 
 Having already investigated with every care the structure of 
 the central polypidom, while we know that in the actual integu- 
 ment of recent zoophytes very peculiar characters generally re- 
 side, it seems most proper to come back, for the purpose just 
 indicated, to the more careful examination of that which has 
 been hitherto only cursorily mentioned, namely, what I have 
 called the polyp-skin. But I have already stated how rarely 
 this exists in any degree of perfection : hence this investigation 
 is attended with great difficulties. And in truth the task is one 
 so difficult that I have frequently been on the point of relinquish- 
 ing all hope of being able to effect it, when the observation 
 of some character which had theretofore escaped attention has 
 renewed my hopes. But for the slitting-wheel and the micro- 
 scope, coupled with a careful study of the matrix, those cha- 
 racters must ever have escaped me. 
 
 It will be convenient, and not in strict language incorrect, if we 
 distinguish the parts of which we have now principally to treat, 
 as the epidermis, cuticle, or polyp-skin already mentioned, and the 
 dermis, cutis, or under-skin, being the delicate membrane already 
 described as immediately underlying the polyp-skin. 
 
 It will be remembered that, in the vast majority of specimens, 
 both in chalk and flint, the central framework of the polypidom 
 is alone preserved f. Both polyp-skin and under-skin, owing to 
 
 * In addition to the points before named, it seems obvious that the con- 
 tractile theory requires that the animal should be strictly a single and not a 
 composite one. It cannot be conceived that a polypiferous animal should 
 have its polypiferous surface contractile over itself. 
 
 t As Tubulipora patina would be, stripped of its cells, and as it is actually 
 described by Mr. Thompson. See Johnston's Zoophytes, p. 268. 
 
32 
 
 their delicacy and the difficulty of extraction, are usually wanting 
 or very imperfect. But it is these latter that must now be scruti- 
 nized. 
 
 A careful examination makes it clear that the Ventriculidse 
 cannot be considered, as Milne Edwards describes the Berenicea 
 diluviana*, as self-enveloping polyps, one series overlying a 
 range of the empty cells of another. Else we should find speci- 
 mens of the same species having different degrees of thickness, 
 and there would be no distinct under-skin such as I have de- 
 scribed. It is quite clear that these polypidoms are true polypi- 
 doms, the living skeletons of living polyp-masses, and not the 
 remains of dead polyps piled one tier above another. 
 
 The careful examination of a suite of the best-preserved speci- 
 mens in my possession at length revealed to me a character 
 which I immediately felt must prove of the utmost importance 
 in the present question. I observed, in casts of some of the 
 most perfectly preserved specimens, minute perforations extend- 
 ing from the surface which had been in apposition with the 
 fossils inwards into the matrix of the cast. Further observation 
 satisfied me tnat this was no accidental appearance, but that it 
 owed its existence to the original of the fossil itself. I found 
 traces of the same appearance also in the casts of specimens 
 preserved in a fine state in flint in those cases in which the 
 remains of the Ventriculite were found in the condition of an 
 open network only, the phenomena of which have been above 
 explained. 
 
 In order to arrive at any just conclusion from these ob- 
 servations, it was necessary to apply here again all the consider- 
 ations which have been already noticed, as to the nature of the 
 two very different matrices in which the Ventriculidse are found. 
 Those considerations showed it to have been a necessary con- 
 dition that the substance, whatever it was, which gave ori- 
 gin to these perforations, was neither of the merely soft and 
 readily decomposable nature of the substance which filled the 
 central tissue of the polypidom, nor yet either composed of a mi- 
 nutely fibrous membrane like the fibrous parts of that polypidom, 
 or being a single projecting fibre overspread with soft or mucila- 
 ginous matter, like a sponge. Were it the former, its place 
 would be most usually filled up, in chalk specimens, by chalk, 
 as the intermediate spaces of the squares have become filled up. 
 Were it either of the latter, we should sometimes find in the flint 
 traces of the fibres hermetically sealed, and in flint and chalk we 
 should find the tubular incrustments and other conditions 
 already stated as marking the former presence of the central 
 
 * M6moire sur les Crisies, &c., Ann. des Sciences Nat. 1838, p. 228. 
 
33 
 
 fibre. But none of such conditions has been ever found. It follows 
 that a substance differing from either formerly filled these minute 
 perforations, a substance little if at all less durable than the 
 skeleton fibre, far more durable than the mere soft intermediate 
 substance. Such a substance precisely answers to the terms 
 under which a true epidermis is properly described, which is not 
 fibrous in texture but pretty nearly homogeneous, and so strong 
 as to have been described by a well-known modern author as 
 " resisting suppuration, maceration, and other modes of destruc- 
 tion, for a great length of time*." 
 
 I have already stated that I had previously satisfied myself of 
 the existence of a distinct polyp-skin or epidermis. A peculiar 
 appearance of corrugation, very difficult to describe but always 
 present in good casts, and with a vacant space between it and the 
 actual fossil, left this beyond a doubt, though at that time no spe- 
 cimen had been found showing the remains of the epidermis 
 itself preserved on the actual fossil. In specimens since found 
 with remains of epidermis preserved, there is, of course, no such 
 vacant space. Attached to this epidermis and of a similar nature, 
 and at least as durablef as itself, there evidently existed during 
 the life of the animal hair-like processes scattered over a large 
 part of its surface. 
 
 Plate VIII. figs. 4 and 5 show examples of the appearances of 
 casts of two very different species of Ventriculidse, in each of which 
 these perforations are present. It will be seen that a different 
 arrangement of them exists in the two cases. 
 
 It would be only tedious to the reader to carry him through 
 the whole process of induction, by which, not through any hasty 
 conclusion, but as the result of careful and very cautious inves- 
 tigation, the reality seemed to force itself upon me that the 
 sources of these perforations corresponded to those curious move- 
 able processes which exist on the surface of some few of the 
 higher zoophytes, and which have been well described by Prof. 
 ReidJ. The determination may seem obvious enough on its an- 
 
 * Elliotson's ' Physiology,' 5th ed. p. 270 ; and see Todd and Bowman's 
 ' Anatomy and Physiology, 5 vol. i. p. 414. The appearance of the polyp- 
 skin, as preserved by the deposit in it of sulphuret of iron, and examined 
 under the microscope, has been recognized by some of my friends wholly 
 unfamiliar with these or other fossils as bearing a striking resemblance to 
 the human epidermis examined in the same way. 
 
 f I have found by actual experiment that, at least in some species, the 
 hair-like processes of recent polypifers are, though hollow, less easily de- 
 structible by maceration in llq. potassce than are the other parts of the epi- 
 dermis. And I cannot avoid noticing here, that on such maceration a gra- 
 nular appearance was assumed very similar to that of the Ventriculidse. It 
 is unnecessary to dwell on this effect of the potash contained in the same 
 felspar rocks to whose disintegration the siliceous fluid has been ascribed. 
 
 I Ann. and Mag. Nat. Hist. vol. xvi. p. 385, &c. 
 
 C 
 
34 
 
 noun cement thus in a few words, but it was not so easy to be 
 made. 
 
 After I had satisfied myself as far as I was thus able upon this 
 point, I determined, if possible, to test the correctness of the 
 conclusion. Taking, therefore, a very fine specimen of Membra- 
 nipora pilosa*, in which the processes, exhibiting every character 
 described by Prof. Reid, were well-preserved, I immersed its face 
 in very finely prepared plaster of paris. On removing it after 
 the plaster had hardened, I found, under the microscope, an 
 appearance identical, as to the perforations, with that presented 
 by the casts of the Ventriculidse. 
 
 The reader who may search for these remains of the movcable 
 processes on the Ventriculidse, and, as many will, may not find 
 them, and who may therefore too hastily doubt the accuracy of 
 my observations, will do well to refer to p. 332-3 of Johnston's 
 ' Zoophytes/ where he will find that, in even recent specimens, 
 " they are not present in every specimen of any of these species, 
 and indeed are very rarely to be seen on some of them, and when 
 present it is only upon some of the cells." If this is the case in 
 recent specimens how much more is it to be expected in fossil 
 specimens that the traces of the former existence of these pro- 
 cesses will be sometimes few, sometimes altogether absent ! The 
 universal friableness of the chalk and the general solidification 
 of the flint often almost or totally obscure them. Doubtless, 
 moreover, the originals of many of the specimens which remain 
 both in chalk and flint were dead when enveloped, for the views 
 already stated require only that the soft parts should have been 
 yet undecomposed, not that the animal should have been alive. 
 
 I am enabled, by some specimens in an extraordinary state of 
 preservation with which frequent personal excursions into the 
 field have rewarded my careful search, to add some description of 
 these processes as existing in the living Ventriculidse and derived 
 from other observation than that of mere casts only. When the. 
 places of all are preserved, which is very rarely the case, they are 
 disposed not without some regularity: it would appear that one was 
 appropriated to each polyp, whose celFs mouth it doubtless swept 
 in the living state as the mouths of the polyp-cells of the Cellu- 
 laria described by Prof. Reid are swept by the processes existing 
 in those species. As in those cases, the process was affixed to a 
 slight projection of the polypidom on the outer edge of each 
 polyp-cell. As in those cases also the processes tapered off gra- 
 dually to a point in the Ventriculida3, though the base of the 
 
 * In this specimen there are all the parts described by Prof. Reid as ap- 
 purtenant to these processes, though Dr. Johnston does not describe the spe- 
 cies as exhibiting them. 
 
process was rather proportionally broader in the latter than in 
 the former. 
 
 Any special peculiarities in these processes exhibited by any 
 species will be noticed in the descriptions hereafter to be given 
 of each separate species. 
 
 The discovery of these moveable processes was certainly an im- 
 portant and very interesting point in the anatomy, as it was an 
 essential aid towards determining the proper position and true 
 affinities, of the animal to which they belonged. It will be ob- 
 served that these processes are affixed to the strict polypidom, not 
 to any distinct and separable external cell, thus adding another 
 to the many evidences afforded by the Ventriculidse that the 
 whole of every polyzoic polyp-mass is a true entire animal, of 
 which the polypidom forms but one essential and inseparable 
 part. 
 
 The last observation leads naturally to the inquiry if the 
 moveable processes are not attached to any distinct and separable 
 external polyp-cells*, as in Membranipora, Eschara, &c., have 
 any traces of the existence and position of the polyps been dis- 
 covered ? Now it is undoubtedly the fact that the surface of the 
 vast majority of these fossils exhibits no trace of polyp-cells. 
 And remembering that many of the most highly organized of 
 the recent Polyzoaf present very faint external traces of cells 
 
 * The separateness, externally, of the polyp-cells is a matter more appa- 
 rent than real. The Ventriculite cells were quite as individually separate, 
 the independence of each other, therefore, of their inhabiting polyps at least as 
 great, as in Eschara, FLustra, &c. ; indeed more so, there being in the former 
 a distinct extent of the under skin between each. In the Halodactylus dia- 
 phanus the cells merely assume a hexagonal form " from their pressure upon 
 each other." This is very inferior to the Ventriculidoe : see Farre's paper 
 above-cited, p. 409, and above, p. 29. Space forbids me to enter so fully into 
 this subject as I could wish, and as its importance might render desirable 
 were it not that the line of argument would be thereby interrupted. I will 
 do no more now than recall the reader's attention to the fact of the organized 
 external integument common to the compound Ascidians, which the under- 
 skin of the Ventriculidse certainly resembles more than does the connecting 
 medium of the Eschara} and to two quotations from Dr. Johnston (Zoo- 
 phytes, p. 255 257) bearing directly on the point. "The cell," he says, 
 " is in fact the outer tunic of the polyp analogous to the envelope of the 
 compound mollusca, * * in organic connexion with the interior parts, and 
 liable to organic changes;" and in distinguishing by a marked character the 
 Polyzoa from the Anthozoa, he says, " In the latter the polyps are simply 
 developments of the common central fleshy mass, identical with it in struc- 
 ture and texture; in the former each individual is a distinct organism, and 
 the medium which binds them together, whether vascular or Hgamentous, 
 has its own peculiar character." If this is correct, there can be no doubt 
 that the Ventriculidse answer to the second description and not to the 
 first, and that indeed in a much more marked manner than do Eschara, 
 Flustra, &c. 
 
 f I use the term Polyzoa in preference to Bryozoa, first, as having a prior 
 
 c2 
 
36 
 
 when dead, I had little hope of ever finding absolute evidence of 
 the presence of those cells in the Ventriculidse. Even in the 
 Escharidse they are often obliterated, while in the Halodactylus 
 diaphanus it is rarely that they can be traced in specimens pre- 
 served in spirits. Even in the recent animal they can only be 
 detected by aid of the microscope. I was then fully prepared to 
 find this part of my researches unsuccessful. 
 
 A very minute examination, however, of my most delicately 
 preserved specimens at length threw some light even on this 
 point. I found that, in all the most marked varieties of form of 
 the Ventriculidse, in that membrane which I have described as 
 lying external to the central part of the polypidom and imme- 
 diately underneath the epidermis, and which membrane is above 
 distinguished as the under-skin, there were numerous spots, scat- 
 tered over the surface, which that membrane did not fill. The 
 size of these, though varying, as was to be expected, in different 
 species, does not do so very greatly, and may be stated as ranging 
 between the 100th and the 200th of an inch. These are arranged 
 in some species with more, in others with less regularity. A very 
 rigorous examination of such specimens, tested by sections, led 
 me, after a length of time, to the conclusion, that these vacuities 
 are the true polyp-cells*. I may observe, first, that they are 
 found in the most different species, including the so-called 
 Ocellaria ; secondly, that they correspond in mode of position 
 and character to the polyp-cells of many of the Polyzoa and espe- 
 cially of the Halodactylus diaphanus, differing only from the latter 
 in being circular, because implanted in a more strongly developed 
 and highly organized connecting medium or tunic which is found 
 regularly separating each one from its neighbourst, a fact con- 
 sistent with the higher degree of organization of these animals, 
 and seeming again to point to the singleness of the entirety of the 
 whole polypiferous mass; thirdly, but not least importantly, that I 
 have since found specimens in which the polyp-skin is, through 
 the deposit of sulphuret of iron as already explained, well-pre- 
 served, in which specimens these cells are seen marked with full 
 as much distinctness in that polyp-skin as in any specimen of 
 Halodactylus diaphanus or of many other recent polypifers. On 
 
 title, second, as having some significance, which Bryozoa has not. It is quite 
 enough to perpetuate error by calling a class animal plants, without in- 
 creasing it by calling one order of that class " moss-animals " ! 
 
 * Jt is important to observe, as again showing the absence of any analogy 
 with the Alcyonhtm, that not in these polyp-cells or on any part of the sur- 
 face of any of the Ventriculidae is there any trace of those stellate figures 
 which distinguish the surface of the Alcyonidse, and indeed of all the Aste- 
 roida, and traces of which would necessarily have been preserved had they 
 or their related objects ever existed. 
 
 f See note, p. 35. 
 
37 
 
 Plate VIII. fig. 6 is a figure of this polyp-skin in a species in which 
 there is considerable regularity in the arrangement of the polyps. 
 
 The polyp-cells are lodged in the substance of the under-skin, 
 extending in the plane of its thickness the whole depth. They 
 lie at distances from one another about equal, or rather more, to 
 their own diameter. In the V. simplex their figure of arrange- 
 ment has a degree of regularity partaking of the quincuncial, a 
 figure so characteristic of the Polyzoa. In the convoluted varie- 
 ties it is (perhaps apparently only) less generally regular. 
 
 What gave a vastly greater importance to and safety in reliance 
 on the facts thus observed, both as to the polyp-cells and the 
 moveable processes, was this : that I found that each of them was 
 only present in certain cases, always absent in others ; and that 
 those places where both were always absent were constant*. 
 Neither polyp-cell nor process is ever found on any part of the 
 root ; nor in those varieties characterized by a head, to which I 
 have already alluded, is either ever found on any part of that 
 headf. It will be felt to be necessarily consistent with the 
 character which I have shown to mark the roots of the Ventri- 
 culidaB that neither polyps nor processes should be found there. 
 I consider it to be the characteristic of the head-bearing Ven- 
 triculidse that those heads are apolypous. The polyp-skin and 
 under-skin are perfect on those heads, exactly the same in all re- 
 spects, excepting in the entire absence of polyp-cells and processes, 
 as the polyp-skin and under-skin covering the rest of the body. 
 The examination of those heads serves therefore as an important 
 test. 
 
 Every fact observed in less perfect specimens is consistent with 
 the important results thus obtained. 
 
 By a section most happily made, I have obtained (PL VII. 
 fig. 13) a transverse view of a few polyp- cells in flint J. The 
 section is somewhat oblique, which it is almost impossible in such 
 minute structure to avoid, and hence the fibres appear irregular 
 owing to those which lie in really different planes striking the eye 
 in the same focus. A slight variation of the focus shows that 
 the - under-skin is closely attached to the polyp-skin, rising 
 
 * I have sometimes thought that I could detect obscure traces of cells under 
 the upper part of a root on the body of the Ventriculite. This would be 
 quite consistent with well-known phsenomena, and I will only refer to Milne 
 Edwards' paper, already cited, p. 211, for a very apt illustration of this 
 obliteration of cells and overgrowth of root-fibre. 
 
 f The existence of this head-bearing group is another strong evidence of 
 the unity and entirety of the whole polypiferous mass. 
 
 J This section was obtained long before I had satisfied myself of the ex- 
 istence and character of the polyp-cells ; and it was only when I had satisfied 
 myself of those facts by other means that I re-examined this specimen and 
 found its entire consistency with those inferences. 
 
38 
 
 where it rises, depressed where it is depressed,, but without any 
 distortion. The same variation shows, in perspective, other ele- 
 vations than those most forward, lying beyond them, as faintly 
 indicated in the engraving. 
 
 I conceive that the processes w T hich I have mentioned, and 
 which had been lost in this specimen before its envelopment, had 
 their places upon the more elevated parts seen in the figure. If 
 the reader will refer either to an actual specimen of Halodactylus 
 diaphanus or hirsutus, or to the figures of the former in Dr. Farre's 
 paper, pi. 25. figs. 2 and 3, he will perceive that the actual place 
 of protrusion of the polyp itself is not more plainly seen in these 
 recent specimens than in the specimen which I am describing; 
 and the peculiar mode of drawing-in the body which charac- 
 terizes this order at once explains the reason of this apparent ob- 
 literation of the cell's mouth. I apprehend that the preservation 
 of this fragment is owing to the circumstance of its having been 
 suddenly torn from a living animal, the portion of polypidom 
 attached to it having lost its soft contents in, or soon after, the 
 rupture, and the torn fragment having thus become hermetically 
 sealed up, according to the process already described, in solid 
 flint. Such is the actual condition in which it is found, and in 
 no other condition or manner can I conceive it probable that any 
 of the polyp-skin itself would be preserved in flint*. 
 
 It will be understood from what has preceded, that in the 
 Ventriculidse the entire exposed surface, external as well as in- 
 ternal, except the roots and the heads of the head-bearing species, 
 were covered w r ith polyps. 
 
 Another point remains to be noticed in which the Ventriculidse 
 are found to agree with the Halodactylus hirsutus. 
 
 In describing the Alcyonidium hirsutum (Halodactylus hirsutus), 
 Dr. Johnston (p. 361) characterizes it as "marked with numerous 
 yellowish circular spots irregularly disposed -" and informs us that 
 " these yellowish spots are produced by clusters of ova lying im- 
 bedded in the cellular texture." The ovarian cells of the Ventri- 
 culid3 might best be briefly described in exactly the same terms ; 
 and this mode of development of the ova is another illustration 
 of the singleness and entirety of the whole polyp-mass f. The 
 
 * It appears obvious that from the very nature of the epidermis it would 
 never be likely to be hermetically sealed up like an isolated fibre since, 
 though inclosed, there would be communication between its parts, which, 
 over such a continued extent of surface, would not be likely to be entirely 
 destroyed. Hence, like fibre which preserved its connexion, it would in 
 time decay away, and either leave a hollow or be subsequently filled up by 
 calcedony. See ante, p. 13. The epidermis would readily separate after 
 death, but any fragments of it in this state would form but an indistin- 
 guishable mass whether preserved in chalk or flint. 
 
 f I had prepared an extensive series of notes illustrative of the general 
 
39. 
 
 ovarian cells in the Ventriculidse lie imbedded in the substance 
 of the polypidom, replacing, where they occur, the substance of 
 the central polypidom, but entirely inclosed, on each surface, by 
 the polyp-skin (PL VIII. fig. 3.) Their character is the same in all 
 varieties. No polyps or processes exist on that part of the polyp- 
 skin which overlies them. It is perhaps riot unworthy to be 
 noticed that there are generally found, on specimens having no 
 ovarian cells, small patches, of the same shape and scattered in 
 the same way over the surface as the ovarian cells, in which there 
 are no polyp-cells. The suggestion naturally occurs, whether 
 these are the spots in which, at the proper season, ova would 
 be developed ? The under-skin is entire in these places. 
 
 Sufficient it is hoped has been adduced to show that the Ven- 
 triculida3 must have been zoophytes of a very high order* ; that, 
 in point of fact, characters are found in them which belong only 
 to the higher among the Polyzoa or Ascidian polyps ; and that 
 those characters belong to different groups among those Ascidian 
 polyps. It has been seen that the moveable processes found in 
 some recent Escharidaj and a few other forms were combined with 
 some important characters found principally marked in the recent 
 Halodactylus ; and it is not unimportant to remark, that the Halo- 
 dactylus is superior to most other recent Polyzoa in not being a 
 mere aggregation of distinct cells, but in having a distinct cen- 
 tral polypidom f, other than the membrane containing its imme- 
 diate polyp-cells, and other than the remains of dead polyp-cells. 
 The central polypidom of the Halodactylus is composed of 
 a network, very irregular however, which the polyps surround 
 on all sides. In this it approaches the Ventriculidse, but the 
 latter far exceeded it in the perfection of all parts of that poly- 
 pidom, as they obviously did also in firmness % as well as in regu- 
 
 analogy between the embryology of the Ventriculidse and that of many 
 recent zoophytes, but I have not embodied them above, fearing that it 
 would tend to complicate the subject and distract attention from the con- 
 nected argument which I have endeavoured to set forth demonstrative of the 
 affinities of the Ventriculidae. 
 
 * I use the language here of those who include Eschara and Flustra 
 among zoophytes. If they were true Ascidians, the Ventriculidas were true 
 Ascidians also, only more highly developed. 
 
 t This also marks Tubulipora patina very beautifully, whose general 
 likeness to many Ventriculidae has been already noticed. I have found a 
 remarkably fine specimen in the London clay of a polypifer, folded up after 
 the manner of Eschara foliacea, but with a distinct and very regularly con- 
 structed central polypidom. The species (? genus) is undescribed. My spe- 
 cimen is five inches square and four inches in thickness. 
 
 J It will occur to the reader that the security of Eschara, Flustra, &c. was 
 ensured either by their parasitic growth or the calcareous nature of their 
 polypidom. The Ventriculite, having neither of those characters, had the 
 special provisions exhibited in its polypidom to secure, still more perfectly, 
 the same objects, while the separate and peculiar nature of its root gave to 
 its polypiferous surface a wider range and sphere than either EscJiara or 
 
40 
 
 larity and elegance of form. They thus combined the characters 
 of different members of the highest orders of recent zoophytes 
 with the addition of other and very important characters present 
 in none of the latter. And those characters are peculiarly in- 
 teresting as leading perhaps to some settlement of the long-moot 
 question both as to the nature and mode of growth of polypi - 
 (lonis*. Let the beautiful and delicate tissue of the central po- 
 ly pidom its entire separation and distinctness both from the 
 substance and structure of the root its investing under-skin still 
 again differing in structure be considered, and it must be per- 
 fectly obvious that the polypidom of the Ventriculida? was no 
 transudation or excretion from the body of the polyps ; and it 
 will surely be no less evident that it owed its origin to no mere 
 vegetative principle. In addition to the peculiar character and 
 beautifully regular structure of the whole mass, pervaded as it is 
 with one aspect of admirable unity t, let the special adaptations 
 of each part, as already pointed out, to subserve the ends of the 
 preservation and safety of the whole and of each individual polyp 
 be considered ; and can the conclusion be resisted, that the entire 
 creature, the whole polypidom with its whole array of polyps, 
 formed one peculiar and highly organized animal, many-headed 
 or rather with many separate special organs of nutrition un- 
 doubtedly, but still a unity, an entire animal, of which the poly- 
 pidom was one constituent and essential part, which can no more 
 be properly considered without reference to its polyps, or its 
 
 Fi urf.ra enjoy a not unimportant point. In these very facts the higher 
 rank of the Ventriculite seems indicated. In this case, as in those, the se- 
 curity of the polyps was the object of the polypidom. The more elaborate 
 the provision for securing any end, the higher that end must be conceived to 
 be, for nature never wastes or squanders her resources. 
 
 * An apparent inconsistency is found on this point in Dr. Johnston's 
 valuable work, the Asteroida being described in terms exactly opposite on 
 pp. 140 and 255. 
 
 fin all the observations as to the unity and entirety of the polypiferous 
 mass, the whole groups of Eschara, Flustra, &c. are necessarily included. 
 It maybe matter of discussion whether the individual polyps of these are not 
 separately true Ascidians. There can be no question, a I. any rate, that it is 
 a part of the necessity of their nature to exist always grouped thus, and in- 
 separably united by an organized medium. That organized medium is 
 clearly therefore necessary for the safety and security and existence of each 
 polyp; and whether we look at the mass of an Eschara foliacea or of a 
 V. simplex, the argument is the same. That organized medium being thus 
 necessary for the safety, security, and existence of each individual polyp, or, 
 if you please, Ascidian, the more perfect the provision in it for that indivi- 
 dual safety and security and therefore for the fulfilment of the ends of the 
 individual existence the higher must be the rank of the creature, whether 
 looked at as one entirety or as an aggregation of individuals. What those 
 provisions are in the Ventriculidae has been seen. The octahedral structure 
 is a special and elsewhere unapproached provision for the security of the 
 individual polyps (or Ascidians). 
 
41 
 
 polyps without reference to it, than can the bones of a niamnial 
 be considered without reference to its soft parts, or the soft parts 
 without reference to the bones ? 
 
 Classification. 
 
 I HAVE thus described with some detail the structure which marks 
 a large group of fossils from the chalk, and have, further, endea- 
 voured to show what are the natural affinities of the group thus 
 marked. The only clue has thus been obtained towards arran- 
 ging, in a true and natural classification, those widely varied 
 forms to which, under various secondary modifications, this 
 structure belongs. The few of these which have hitherto been 
 known have been uncharacterized except by names as various as 
 the different writers, and which, being names merely, could leave 
 no impression of reality on the inind of the inquirer. 
 
 It will assist the inquirer, and will much enhance the import- 
 ance of the present investigation, if, before entering on the de- 
 scription of their modifications, something is said of the strati- 
 graphical distribution of these fossils. 
 
 From what has already been stated, it will be obvious that these 
 fossils require to be sought : they can seldom fall in the collec- 
 tor's way as do fossils having solid parts, Testacea, Vertebrata, 
 &c. If found at all in the hands of the dealer they will usually 
 be fragmentary only, or in a matrix, the flint, the deceptive 
 character of whose obvious appearances has been already shown. 
 By far the greater part of the forms assumed are, besides, such that 
 no blow of the hammer can disclose the character of the fossil. 
 
 It is necessary to premise thus much that it may be under- 
 stood that the fact of these fossils not having yet been recognized 
 in particular localities or strata is no proof that they do not exist 
 therein ; and, now that the true structure characteristic of them 
 has been described, it may be hoped that the presence of some 
 representatives of the family may be detected much more widely 
 than has been hitherto suspected. A mere fragment may now 
 serve for the detection of that presence*. 
 
 As far as can be gathered from the various authorities already 
 cited, it would appear that these fossils are more abundant in 
 England than in any other country. In the chalk of Kent, 
 Sussex, Norfolk, Wiltshire, and the respectively adjoining locali- 
 ties, some of the forms are abundant, though in each region the 
 localities in which they abound are certainly restricted. In the 
 chalk of Yorkshire they appear to be much less abundant. In- 
 deed many bodies which have heretofore been grouped as Ventri- 
 culidcB from that region have no relation to that family ; while 
 
 * Of course not for the determination of species, or, necessarily, even of 
 genus. 
 
42 
 
 the forms hitherto collected there of true Ventriculidte are very 
 rare, if we may judge from the specimens in the museum of the 
 Yorkshire Philosophical Institution, for the opportunity of care- 
 fully inspecting which specimens I am indebted to the courtesy 
 of Mr. Charlesworth*. In England these fossils have not hitherto 
 been recognized in any other than the Cretaceous group. It is 
 probable that careful search will reveal them throughout all the 
 members of that group. At present they have been found in 
 five divisions of it ; viz. the Upper Chalk, the Middle Chalk, the 
 Lower Chalk, the Chalk Marl, and the Upper Greensand. The 
 prevalence indeed of certain forms is characteristic of certain of 
 these divisions f ; a result which unexpectedly displayed itself 
 after the classification presently to be exhibited had been worked 
 out from a cautious study of the individuals, and the value of 
 which result must therefore strike every inquirer. The particular 
 divisions characterized by the predominance of one or the other 
 class of forms will be shown in severally describing those forms. 
 
 There is no a priori reason why representatives of these forms 
 should not be found in older and in newer formations than these 
 cretaceous beds. Still the fact of their not having been thus found 
 in England, where, in those cretaceous beds, some of them so much 
 abound, leads to some hesitation in relying implicitly on the 
 alleged much lower stratigraphical position of some foreign forms. 
 The foreign forms from the true chalk appear to be few r and rare ; 
 but there are several figures in Goldfuss, to some of which I have 
 already alluded J, probably representing forms belonging to this 
 family, which are there given as from the " Jurakalk." In the 
 
 * In addition to the acknowledgements which I have already made, I 
 have the further pleasure of now recording the kindness, in affording me 
 the means of examining different specimens from very various localities, of 
 Mr. Lyell, Mr. Wetherell and Mr. Oakeshott of Highgate, Mr. Cunnington 
 of Devizes, Mr. Catt of Brighton, and Mr. Whittle of St. John's College, 
 Cambridge, besides that of Mr. Charlesworth as above-mentioned. I must 
 also acknowledge the kind assistance afforded me by Mr. Waterhouse of 
 the British Museum, in facilitating the task of inspecting the specimens in 
 that collection. To the President of the Geological Society I am also in- 
 debted for the prompt courtesy with which he has enabled me to avail 
 myself efficiently of illustrations from the valuable museum of that Society. 
 
 I would take this opportunity of saying that I shall be greatly obliged 
 by any illustrations and opportunities similar to those which have already 
 been so kindly and liberally afforded to me. 
 
 f See observations in the Ann. and Mag. of Nat. Hist. vol. xx. p. 337. 
 An interest beyond even that which they are calculated intrinsically to ex- 
 cite is thus given to these fossils, of the same nature as that which attaches 
 to a series of Ammonites from different beds. 
 
 J Ante, p. 6. It is proper to state that there are many forms, besides 
 those thus specified, figured by Goldfuss as from the Jurakalk, and which 
 I have not much doubt are Ventriculidae. I have only enumerated the 
 more obvious. 
 
43 
 
 British Museum, again, is a large and valuable series of fossils, 
 which I have carefully examined, and which I can therefore state 
 with assured confidence to belong to this family *, and which are 
 stated to be from Mount Ehanden in Switzerland ; a locality the 
 strata of which are declared to be equivalent to the lower beds of 
 the Middle Oolite of England. The matrix appears much the 
 same as that of our English chalk marl ; but that test is, of course, 
 very incomplete. The point requires careful investigation ; and, 
 as the true character and importance of these fossils will have 
 now become known, it may be hoped that the attention of some 
 of the many competent foreign observers may be directed to it. 
 As I shall show the changes which these forms have undergone 
 in passing from one division of our English strata to another to 
 have been great, it will be peculiarly interesting to ascertain 
 exactly to what strata these foreign forms do actually belong ; for 
 many of them differ much from our English forms. It is inter- 
 esting at present to remark that the form which is of the great- 
 est vertical range in the English beds (Brachiolites digitatus] is 
 unequivocally found in these Rhanden beds. 
 
 On p. 510 of the first volume of the Journal of the Geological 
 Society there is described by Mr. Lonsdale, under the name of 
 " Ocellaria ramosa" a fossil found by Mr. Lyell in the Eocene 
 deposits at Jacksonborough in Georgia, United States. Did this 
 fossil exhibit any true affinities with the group which has been 
 called Ocellaria it would necessarily belong to the Ventriculidse, 
 and I was anxious to ascertain the facts. Mr. Lyell has obli- 
 gingly enabled me to do this by placing in my hands all the 
 specimens found by him, and which are, it is believed, all that 
 have ever been found. The result is, that the fossil is found to 
 present none whatever of the characters of Ocellaria ; and I 
 cannot understand upon what grounds it has had this name affixed 
 to it by Mr. Lonsdale, except that he appears, from his observa- 
 tions, never to have had an opportunity of examining any actual 
 specimens of the so-called Ocellaria, and to have been misled by 
 some of the figures f. These fossils however answer to no part 
 of the generic description given by Ramond, or any subsequent 
 writer, of the Ocellaria. The tubules in the Eocene fossils are 
 tubules ramifying through a massive substance, and there is not 
 any polyparium which is " explanato-memlranaceum" and " utro- 
 
 * These treasures are at present un-arranged. I should be happy to 
 assist in that task, and to complete it by adding, as far as possible from 
 my private collection, all the British forms, should the present Commis- 
 sion result in any prospect of improvement in that department. 
 
 t The figure specially referred to, and which is copied by Lamouroux, 
 pi. 72, fig. 5, has certainly a considerable resemblance to a special fractured 
 surface of the Eocene fossil. 
 
44 
 
 que latere porosum." The characters of Ocellaria, as given by 
 all the authors*, are clear and unmistakeable so far as they go; 
 and there cannot be a moment's doubt as to what the true rela- 
 tions of the so-called genus are, as will presently be seen. It is 
 perfectly certain that this so-called Ocellaria ramosa^ has none 
 of these relations, and therefore that it does not serve to bring 
 the Ventriculidse within the tertiary period. No trace of this 
 family has, then, yet been anywhere discovered higher than the 
 upper beds of the English chalk. 
 
 As it is desirable to have the treatment of the subject as com- 
 plete as possible, so far as it goes, and as the materials which I 
 have collected from the English chalk are sufficiently abundant 
 to lead me to hope that such completeness may be given, for all 
 practical purposes, to the description of the forms found in those 
 beds, I shall confine myself at present to these last ; which I the 
 rather do in that, while it would be a priori probable that the 
 examination of so extensive a series of beds would at any rate 
 afford a full series of typical characters, and therefore a sound 
 basis for a permanent and generally applicable system of classifi- 
 cation, the examination of the Rhanden specimens in the British 
 Museum has satisfied me that all of them will range within the 
 typical groups which the forms of the English chalk have led me 
 to assign. 
 
 I have already indicated { in what direction we must look for 
 the essential characters which mark this whole family. It is 
 extremely improbable that a structure so extraordinary, so pecu- 
 liarly bearing the marks of special design and adaptation as 
 the octahedral structure, should be otherwise than characteristic 
 of the family in individuals of which its existence has been dis- 
 covered. Until, then, it has been found elsewhere, the philoso- 
 phical inquirer will take that structure as his guide in the deter- 
 
 * Those characters are, " Polypier pierreux, aplati en membrane, diverse- 
 ment contourne, subinfundibuliforme'e, asuperficie arenacee, muni de pores 
 sur les deux faces." The observations of Milne-Edwards, on an inspec- 
 tion of the actual fossils, are alone sufficient to show that the present fossils 
 could not be Ocellaria, their apparent tubules being, as stated on p. 511, 
 sometimes penetrated by fibres in a radiated manner. Milne-Edwards 
 expressly says (Lamarck, Anim. sans Vert. ii. p. 291), " L'axe solide, qui 
 remplit assez ordinairement les trous, et qui a etc pris pour une partie du 
 Polypier lui-meme, n'est que la gangue qui s'est moulee dans ces trous, et 
 qui s'est cassee au niveau de la surface du Polypier, lorsque celui-ci a ete 
 detache de la masse qui le renfermait." 
 
 f The fossil is however a very curious and interesting one. Its whole 
 aspect and character recall those of the Alcyonium, both in its massiveness, 
 its cylindrical tubules, and their connecting plexus of fibres. I have many 
 analogous fossils from the chalk, into the investigation of which it is my 
 intention to enter when the present subject shall be completed. 
 
 I Ante, p. 31. 
 
45 
 
 mination of the members of this family. My careful attention 
 has therefore been directed to ascertaining the presence of that 
 structure under every various mask of external form, and- 1 have 
 hitherto invariably found that presence accompanied by certain 
 other characteristics, which would necessarily be present if the 
 affinities which I have already attempted to show are those of 
 the Ventriculidse be the true ones. Without full confidence in 
 the Law of Unity as a sure guide, I cannot conceive of any pro- 
 gress being made in any scientific investigation. I have not 
 found that guide to fail me yet in the present investigation, and 
 am therefore content to take it as the basis of such exposition as 
 I am now able to give of the genera and species of the family 
 VENTRICULID.E. 
 
 Proceeding therefore on this basis, it may be stated generally, 
 that all those fossils which are marked by a membranous struc- 
 ture made up of cubic squares, with equally subtending octahedral 
 fibre at the angles of union of those squares, belong to the family 
 Ventriculidse, and that all members of that family are marked by 
 that structure. We shall find, it is true, thus associated, forms 
 externally most diverse*, and the alleged affinity of which would 
 at first sight startle the inquirer ; which have indeed hitherto had 
 places the most different assigned to them : but I shall be able to 
 show that other and most interesting Unities prevail through all 
 these various forms in addition to that structural one ; and these 
 diversities will thus become only another useful addition to the 
 often repeated but too often neglected lesson, that no guide is 
 more fallacious than likeness or unlikeness of mere external 
 formf. 
 
 " A natural classification," says Milne-Edwards, " is nothing 
 else than a description of the modifications, more or less import- 
 ant, observed in the structure of animals, and a specification of 
 the differing degrees of likeness or unlikeness which the latter 
 bear to each other J." Nothing is easier than the multiplication 
 of genera and species. But it is no slight task, though a most 
 important one, to determine what are the material modifications 
 on which distinction of genus should be founded ; what the ma- 
 
 * On the other hand, I shall take a future opportunity of showing that 
 forms externally bearing much resemblance to the Ventriculidse have in 
 truth a very different structure and affinities. 
 
 t Parkinson long ago remarked, that " if the figure of the fossil be as- 
 sumed as the leading character of the species, substances, differing mate- 
 rially in their structure, will be classed together in the same species ; and, 
 on the other hand, if the species be formed on the external structure, we 
 shall have under the same species substances differing widely in their 
 forms." Vol. ii. p. 128. It would have been well if Goldfuss and others 
 had paid a little attention to these important truths. 
 
 } Sur les Crisies, &c., p. 233. 
 
46 
 
 terial points of likeness or unlikeness which should mark separate 
 species*. 
 
 The only principle upon which I can understand any philoso- 
 phical or natural classification to be founded, is the taking some 
 principal and most easily recognizable point in the oeconomy of the 
 living animal, and examining all the individuals under review in 
 reference to that one point. Can we apply this test in the present 
 case ? It has been already seen that the Ventriculidse belong to a 
 high type of the Molluscan Polyps, to the Polyzoa, approach- 
 ing most nearly to the recent Eschara and Plalodactylus. The 
 fossilized remains of animals of this order, the organization of 
 whose recent congeners has been but so lately understood, might 
 seem at first sight to baffle any attempt to seize on such a point. 
 It seems to me however that such a one may be found. In all 
 recent animals of this order the first essential to their life and 
 well-being is the presence and free access of the sea-water. Va- 
 rious contrivances are adopted to secure this end, some genera 
 and species being parasitical, some loosely floating, some stiffly 
 erect ; each, varying as they also do in form, adapted to the pe- 
 culiar circumstances of the locality which it inhabits, and each, 
 according to the particular plan adopted, exhibiting some charac- 
 teristic differences in habit and organs. This is precisely consistent 
 with the observations already madef as to the constant relation 
 existing between the polypidom, rightly examined, and the nature 
 of the inhabiting, polyps. Such differences no doubt existed in the 
 recent Ventriculidae ; and though it is obviously impossible that 
 we should ever be able, in these fossils, to ascertain the points of 
 difference in habits and individual organs, we may, by care and 
 patience, ascertain those differences in the contrivances displayed 
 in the structure of the polypidoms which we must thus be satis- 
 fied were intimately and necessarily connected with such differ- 
 ences in habits and individual organs. I allude to the various 
 modes of folding of the delicate membrane J which forms the 
 framework of every individual of this family, and on whose sur- 
 face the minute and numberless colony of polyps dwelt. I ap- 
 
 * Were I to follow the example of some botanists, who, for example, in 
 a favourite tribe, the Cactus, have amused themselves with hair-splitting of 
 genera to a marvellous extent, I might readily succeed in perplexing the 
 inquirer with a great multitude of unintelligible names. Between many 
 of the species which I have grouped together, differences far more marked 
 exist than those by which these gentlemen and too many paleontologists 
 have overlaid the intelligibility of their classifications as generic distinc- 
 tions. 
 
 f Ante, p. 26-28. 
 
 { A membrane, it will be remembered, which, by its structure, was firm 
 like the Eschara (though not calcareous), and not loosely floating like the 
 Halodactylus. This is important in considering the permanence of the 
 different modes of folding adopted. 
 
47 
 
 prehend that it can need no detail of argument or mathematical 
 demonstration to show, that upon the mode and degree of fold- 
 ing of this membrane, the greater or less freedom of access, 
 change, and circulation of the water, and its consequent power 
 of being acted upon by the numerous ciliated tentacles and move- 
 able processes, must have depended. Every one who is familiar 
 with the difference in mere circulation of air between the narrow 
 street and the open road, between the deep valley and the hill 
 top, will recognise the essential importance in this respect of every 
 difference in that mode and degree of folding ; and, when the 
 extreme minuteness of the individuals is considered, it will appear 
 that variations of fold hardly appreciable to the eye will have 
 probably had a material influence on the condition of the tenants 
 of these wonderful structures. I cannot doubt that every con- 
 stant difference in the mode and degree of folding of the Ventri- 
 culitic membrane was accompanied by some modification in the 
 organs or habits of the animals, adapting them to that particular 
 mode and degree of access, change, and circulation of sea-water 
 which that mode and degree of folding made a matter of absolute 
 necessity*. 
 
 Taking then the Ventriculidse as a family of the Polyzoa, I 
 shall first endeavour to show that there are certain broad and 
 very marked constant modifications in the mode of foldingf cha- 
 racterizing certain extensive groups which yet have many points 
 of constant difference between the individuals which, as groups, 
 are respectively thus characterized. These groups will form 
 distinct genera. I shall show that certain subordinate but yet 
 important modifications mark, in common, several of the indivi- 
 duals of each of these genera, which individuals yet have further 
 still subordinate but constant and therefore characteristic points 
 
 * Sir J. G. Dalyell, in his recent work on ' Remarkable Animals of Scot- 
 land/ especially notices the importance of attention to the varying condition 
 of the water in which specimens are kept as the great secret of their preserva- 
 tion ; and even his care has often failed. A " low organization " and slight 
 sensibility have been hastily attributed to Polyzoa from their enduring great 
 changes of heat and cold. There is no animal capable of enduring greater 
 changes in this respect than man. But take another class, and it is well- 
 known that from the same heap of frozen fish one may be dashed to shivers 
 on the ground, while another, put into a pail of water, will, in two minutes, 
 be swimming about. 
 
 f* The inquirer will at once perceive the difference between this and mere 
 external form. The same general external form may mask numberless most 
 different modes of folding. My object is to aid in realizing, by classifica- 
 tion, the living animal in all its integrity and varieties. By the accumu- 
 lated names Scyphia, Coscinopora, Guettardia, &c., nothing ever was or can 
 be vivified ; no real idea conveyed to the mind. But the object of the natu- 
 ralist should surely be, not an accumulation of mere names, but the realiza- 
 tion of living and true ideas of various absolute modes of actual existence, 
 be they past or present. 
 
48 
 
 of difference. The inquirer is thus further relieved from the 
 detail of specific differences by the division of each genus into 
 sections. The still subordinate but constant points of difference 
 last named will be characteristic of species. 
 
 I have already alluded to the important and valuable test of 
 the soundness of these principles of classification afforded, un- 
 expectedly and after the work was completed, by the stratigra- 
 phical harmony exhibited by the table of classification. It will 
 be sufficiently obvious that the ocean of different ages would 
 have such modifications as would not be adapted equally to all 
 varieties. We accordingly find among the Ventriculidse, as in 
 other divisions of palaeontology, a few species enduring through 
 many changes; others dying out; while with every fresh sera 
 fresh forms display themselves. 
 
 It will be understood from this, that mere size does not enter 
 as an element into the determination of genus or species. Of 
 many species I have specimens from an inch to eight or nine 
 inches in diameter. It is not necessary to enter very fully, there- 
 fore, into the question of growth. That question, always a difficult 
 one in paleontology, is difficult even in recent forms of the families 
 allied to the Ventriculidse. It would be vain to hope to throw 
 much light upon it by fossil forms. Where constant differences 
 are found under all varieties of size, we are bound to consider 
 them as distinct species. I shall touch briefly on the question 
 of growth in introducing each separate genus. 
 
 It will be also understood that the mere external (outward or 
 inward) general form of the fossil does not enter as an element 
 into the determination of genus or species. I have shown how 
 deceptive that criterion must ever be. In the present instance 
 the same general external form conceals essential differences in 
 the mode and degree of folding of the membrane. 
 
 It will occur to the reader that to follow the fold of a mem- 
 brane, the trace of which is preserved only in a hard and solid 
 matrix, must be a work of great difficulty ; and especially when 
 that matrix is either so friable as the chalk, or so impracticable as 
 the flint. The actual amount of the difficulty * cannot however be 
 fully appreciated without actual experiment. The presence of that 
 very oxide of iron, without which the forms could not be, in general, 
 
 * In order that the actual nature, importance, and results of the present 
 investigation should he properly understood, it is necessary to remind the 
 reader that from the time of Dr. Mantell's first work to his latest, and 
 either by him or the other latest writers (see Portlock's 'Report, &c.'p. 342), 
 it has never been suggested or suspected that any membrane whatever 
 existed in any of the Ventriculidse. They all describe them as composed 
 of anastomosing " cylindrical fibres," (see ante, p. 4,) between which, on 
 the inside, papillae or tubuli arise. I have demonstrated that the basis of 
 the Ventriculidse is a simple unperforated membrane ; that, therefore, the 
 
49 
 
 even detected, necessarily stains the matrix beyond the structure 
 itself; and it requires the nicest and most painful discrimination 
 to determine what is due to structure and what to mere iron 
 stain. Feeling however that such a course of investigation could 
 furnish the" only true materials of a natural classification,, I have 
 endeavoured to overcome these difficulties. And it may save the 
 task both of making and answering many objections if I now 
 state that I have, with this object, dissected with elaborate care 
 numberless specimens, in addition to many hundreds of sections 
 of specimens both in flint and chalk, which, with the like purpose, 
 I have made. There is not one species which I have established 
 which I have not determined from actual and personal section 
 of specimens either in chalk or flint, usually both, and in which, 
 with scarcely an exception, I have not followed and traced out 
 the actual fold with the knife and needle. 
 
 My aim has been to present such a classification and no- 
 menclature as should be intelligible and at the same time ex- 
 pressive; which, whether respect be had to genus, section, or 
 species, should give some accurate and specific idea of the point 
 on which the respective division has been founded ; that thus a 
 mere inspection of the table of classification may carry with it 
 some real and true ideas as to the objects included*. The name 
 
 descriptions so long before the world, and so often repeated, are funda- 
 mentally erroneous, the conclusions as to the oeconomy of the animal being 
 necessarily, therefore, as fundamentally erroneous. It is upon the same 
 laborious care which has enabled me to demonstrate these facts, that I rely 
 in attempting the descriptions now to be given of the different modes of 
 folding assumed by that membrane, and the superficial appearances of 
 which have misled these observers. 
 
 * It is usually unadvisable to alter names once applied ; but where the 
 character of an object has been wholly misunderstood, not even its generic 
 or structural character having been known (see the last note), there can 
 be no claim to retain old names. Their retention is then generally mis- 
 chievous as a mere perpetuation of error. I fully agree with Dr. Farre (ut 
 ante, p. 405, note) that oftentimes " confusion and doubt (in nomenclature) 
 can only be dispelled by beginning de novo," and so applying new names 
 in harmony with a system founded in nature and upon some definite prin- 
 ciple. I think it better to give here all the names which occur in Mr. 
 Morris's Catalogue whose objects appear to belong to the Ventriculidse, 
 a list which will, moreover, show the " confusion and doubt " which have 
 hitherto prevailed in the nomenclature of this family. 
 
 Names in Morris's Catalogue. In the following classification. 
 
 Choanites flexuosus Ventriculites latiplicatus. 
 
 Choanites subrotundus Cephalites constrictus. 
 
 Ventriculites alcyonides [OcellariaJ Ventriculites quincuncialis. 
 
 alternans Probably V. bicomplicatus. 
 
 Bennettise One of the Cephalites annulati, but 
 
 no accurate description ; and the 
 figures of Michelin and Manteil 
 
 (V. infundibuliformis) totally differ. (V. cavatus) 
 
 D 
 
50 
 
 applied to the whole family and to the first genus is the only 
 apparent exception to this rule*. The name Ventriculites would 
 certainly not have been applied to any of these bodies, or to the 
 family, by myself. It was applied by Dr. Mantell to the few 
 forms found by him, under the idea of the internal cavity being 
 the true digestive surface of a single animal. Though the idea 
 under which the name was thus applied has been shown to be 
 wholly erroneous, I have been unwilling, out of respect to the 
 many labours of Dr. Mantell in the field of palaeontology, to 
 reject, as others have done without assigning any reason, this 
 generic appellation ; and I have justified myself in its retention 
 by the classical use of the same word, though in a secondary 
 sense only, in a very different way, viz. as applied to mere sacci- 
 form cavities f. It will be understood, therefore, that the terms 
 Venlriculida and Ventriculites bear no reference to any digestive 
 cavity, but simply to the fact of the creatures to which they are 
 applied always assuming forms which display a central cavity 
 more or less simple. I am glad that this modification in the 
 meaning of the word enables me to retain a name which will 
 always bring to the inquirer's recollection the long and successful 
 labours of Dr. Mantell. 
 
 It is impossible to examine an extensive series of remains ex- 
 hibiting the characteristic structure of the Ventriculidse, without 
 perceiving that, however widely in other respects the individuals 
 differ from one another in the mode of fold of their membrane, 
 they all range themselves within one or the other of three strongly 
 marked and constant modifications, quite independent of mere 
 size. 
 
 The first in natural order, as having most of that simple pouch 
 form which is implied in the name Ventriculida as above ex- 
 
 Names in Morris's Catalogue. In the following classification. 
 
 Ventriculites infundibuliformis Ventriculites cavatus or bicompli- 
 
 catus. 
 
 quadrangularis Brachiolites angularis. 
 
 quadratus ? not a Ventriculid. 
 
 radiatus Ventriculites radiatus. 
 
 Ocellaria inclusa quincuncialis. 
 
 nuda Ibid. 
 
 Spongia Townsendi Ventriculites simplex. 
 
 labyrinthica Brachiolites convolutus. 
 
 Scyphia Fittoni Fragment of Brachiolites digitatus. 
 
 * The termination "ites" is not in itself very classical, but has been so 
 generally employed as to be a convenient and intelligible distinctive mark 
 of fossil generic appellation. Hence I retain it in " Ventriculites," and am 
 therefore obliged so to terminate the other generic names. I am glad to be 
 able to retain, consistently, Dr. Mantell's specific name radiatus. 
 
 f* Thus Cicero : " x ea [anima] pars concipitur cordis parte quadam, 
 quam ventriculum cordis appellant, cui similis alter adjunctus est in quern 
 sanguis a jecore per venam illam cavam influit." De Nat. Deor. ii. 55. 
 
51 
 
 plained, are a large number whose general form is that of a more 
 or less open or close sac, the wall of which rounds or thins off to 
 a marginal edge. All of this kind are single, and supported on 
 a single root, unless in those few abnormal cases before men- 
 tioned*, and which afford no exception to the principle either of 
 the structure or classification. Where, as very rarely occurs, two 
 are united, it is at the roots that they are united. They are 
 not branches of one body. 
 
 All these forms I distinguish by the name of VENTRICULITES. 
 
 Next to these are naturally placed another group, all the mem- 
 bers of which are much rarer than the last, most of them of great 
 rarity, but yet exhibiting a diversity of forms as great, well- 
 marked and constant as the different individuals of the genus 
 Ventriculites. All however are marked by the very striking 
 peculiarity of the wall of the pouch not thinning or rounding 
 off to a marginal edge, but being crowned by a broad and distinct 
 head, prominent and well-defined, and totally differing in aspect, 
 structure, and function from the rest of the body. This charac- 
 teristic suggests, as peculiarly appropriate, the generic appellation 
 of CEPHALITES. 
 
 The two genera thus distinguished each exhibit, though with 
 striking modifications, more or less of the simple pouch form in 
 their internal cavity, or of obvious singleness in the general shape 
 which the fold of the wall of their cavities, or their apolypous 
 head, assumes ; but a large group remains to which neither cha- 
 racter applies, and all the members of which stand out conspi- 
 cuously as folded in many lobes and in many broadly separated 
 parts. The word brachium being often used by the best authors 
 in. the sense of projection simply, I use the diminutive of that word 
 to distinguish all of this group by the name of BRACHIOLITES. 
 
 But, again, the individuals comprised within the description 
 of the genus Ventriculites are found to exhibit two broad modi- 
 fications in the general aspect of the membrane composing the 
 wall of the pouch. The two sides of the wall correspond in the 
 one group, both surfaces being either smooth, or, if marked with 
 folds, the depression of one side having a corresponding eleva- 
 tion on the other; in the other this correspondence is absent, 
 owing to some change in the direction of the fold before reaching 
 the opposite surface, as already alluded tof- It will materially 
 assist the memory and researches of the inquirer if we accord- 
 ingly divide the genus Ventriculites into two sections, which I 
 distinguish by the names Simplices for those species having cor- 
 responding surfaces, Complicati for those which change the direc- 
 tion of their fold between the two surfaces. 
 
 * Ante, p. 18. f dnte, p. 16. 
 
 D2 
 
52 
 
 So the individuals comprised within the description of the 
 genus Cephalites exhibit two broadly-marked modifications ; the 
 head of the one group being only of the same breadth as the thick- 
 ness of the wall, and being placed exactly at the top of that wall, 
 and nearly at right angles, at every point, to the outer and inner 
 
 surfaces of that wall ; the head of the other group being much 
 broader than the thickness of any part of the wall, and never 
 lying flat at the top, but extending more or less down over the 
 sides of the wall. These marked differences are accompanied by 
 important differences in the mode of fold of the membrane. I 
 distinguish therefore the genus Cephalites into the two sections 
 Annulati, being those in which the head extends as a mere broad 
 ring round the flat top of the wall, and Dilatati, being those in 
 which it is spread out so much more extensively. 
 
 And so also the members of the genus Brachiolites are at once 
 separated into two groups, by the remarkable circumstance that 
 some of them have the extremities of those projecting lobes into 
 which they are divided open, others closed. The latter I distin- 
 guish as the sectional division Operti, the former as Aperti. 
 
 I shall hereafter point out the minor modifications accompa- 
 nying these more striking ones, and endeavour to show the final 
 purposes of the respective modifications themselves. 
 
 It will of course be well understood that, as in every class of 
 fossil forms the exact determination of the species of individual 
 specimens is often difficult, frequently impossible, such must 
 sometimes be the case with respect to the Ventriculidse. The con- 
 ditions under which they are found render them peculiarly liable 
 to this difficulty ; and the inexperienced observer who has not 
 yet learned to distinguish that which is a mere cast* from a speci- 
 men in which some of the actual body is preserved, a task of 
 no slight difficulty, and only to be successfully undertaken after 
 acquiring a full knowledge of structure, will often find himself 
 baffled in the attempt at specific identification. Hence the 
 importance of attention to those sectional and generic characters 
 already noticed, and which he will rarely be unable to distinguish. 
 
 These broad modifications, and the respective relations thereto 
 
 * Michelin's Ocellaria grandipora, pi. 40. 3 a & 36, is a mere cast of ex- 
 ternal and internal surfaces. The imperfection and indefiniteness of almost 
 all the figures yet published have been already noticed (ante, p. 6-8). It 
 would therefore be a useless attempt to endeavour to identify them. Ob- 
 jects of this class require to be well understood before they can be truth- 
 fully represented by figures. The figures of Dr. Mantell are no exception 
 to this remark, as they only give the broad external characters of one species 
 (which they however do) without any indication of the mode of fold of the 
 membrane which gives rise to those characters, and the very existence of 
 which membrane Dr. Mantell denies. 
 
53 
 
 of the minor modifications, will be the better understood from 
 the following table of classification, in which I have arranged the 
 species belonging to each genus in such relative position as should 
 best display the transition from one general character of folding 
 to another, and thus gradually realize the true relations existing 
 between the very different forms which lie at the two extremes. 
 
 Class MOLLUSCA TUNICATA *. 
 
 VENTRICULITES, Mant. 
 a. SIMPLICES. 
 
 1. simplex. 
 
 2. impressus. 
 
 3. quincuncialis. 
 
 4. muricatus. 
 
 5. tessellatus. 
 
 6. cavatus. 
 
 7. striatus. 
 
 b. COMPLICATE 
 
 1. mammillaris. 
 
 2. latiplicatus. 
 
 3. decurrens. 
 
 Var* tenuiplicatus. 
 
 4. radiatus, Mantell. 
 
 5. bicomplicatus. 
 
 Order POLYZOA f. 
 Family VENTRICULID.&. 
 
 CEPHALITES. 
 a. ANNULATI. 
 
 1. longitudinalis. 
 
 2. guttatus. 
 
 3. paradoxus. 
 
 4. alternans. 
 
 5. bullatus. 
 
 7. eaten ifer. 
 
 Var. annulatus. 
 
 8. compressus. 
 
 b. DlLATATI. 
 
 1. capitatus. 
 
 2. campanulatus. 
 
 3. constrictus. 
 
 4. perforatus. 
 
 BRACHIOLITES. 
 a. OPERTI. 
 
 1. tuberosus. 
 
 2. elegans. 
 
 3. corivolutus. 
 
 4. angularis. 
 
 b. APERTI. 
 
 1. foliaceus. 
 
 2. racemosus. 
 
 3. digitatus. 
 
 4. tubulatus. 
 
 5. fenestratus. 
 
 6. labrosus. 
 
 7. protensus, 
 
 * It is quite beyond my present purpose to discuss the exact position of 
 the Potyzoa. The main truth of Professor E. Forbes's opinion is however 
 so generally recognized, that I am justified in the above designation of class, 
 which must always be felt to be an important element in giving vitality to 
 a classification. Professor Forbes says, "The anatomical structure of the 
 Ascidioida or Bryozoa removes them altogether from the class of Zoophyta 
 into that ofMollusca, where they should form an order of Mollusca tunicata 
 parallel with the group of compound Tunicata of which Botryllus and such 
 forms are examples/' Ann. and Mag. of Nat. Hist. vol. xiv. p. 390. See 
 Owen's Lect. on Comp. Anat. I. pp. 100 and 269, 270 ; Van Beneden, Re- 
 cherches sur les Bryozaires, p. 37 ; Johnston's British Zoophytes, p. 2. 
 (See also Thompson and Farre.) The actual and important distinctions are 
 noticed by the last writer, p. 256 ; and the vast superiority in vital activity 
 of the Polyzoa to the Ascidians, is well pointed out by Sir'j. G. Dalyell (ut 
 ante) pp. 229, 230. I have heretofore used the general term " zoophytes " 
 in conformity, as already intimated (note, p. 39), to what is at present the 
 ordinary language of authors, and a departure from which would, therefore, 
 have caused ambiguity and unnecessary confusion. [The reader may be 
 usefully referred, hereon, to the earlier pages of Forbes and Hanley's ' Bri- 
 tish Mollusca.'] 
 
 Thompson ; Bryozoa, Ehrenberg ; Ciliobrachiata, Farre. 
 
54 
 
 Dcscrijjtions of species. 
 Family VENTRICULID^. 
 
 Character. Structure. Polypidom membranous : membrane com- 
 posed internally of fibres arranged in several usually five* 
 layers of cubic squares, equal, for the most part, in the plane of 
 thickness and of superficies, and connected at all their angles 
 by other fibres having a regular octahedral arrangement : exte- 
 rior to this, both within and without, a dermis composed of a 
 single layer of smaller squares, and in which the polyps are 
 lodged, usually on both surfaces : exterior to this a simple epi- 
 dermis : roots distinct, less regular in structure and without oc- 
 tahedral fibre. 
 
 Habit. One or more central cavity, the principal opening to 
 which is at the top : roots ensheathing base of polypidom and 
 extending below into radicles ; never affixed to solid bodies. 
 
 The details which have been already given render further ob- 
 servations on the characters of the family unnecessary. 
 
 Genus VENTRICULITES. 
 
 Character. Pouch-shaped; varying greatly in size and dilata- 
 tion : cavity single and regular : membrane forming the wall 
 of the cavity either simple and smooth on both surfaces or 
 more or less closely and regularly folded (thus giving it a ru- 
 gose character) : margin of wall thinned or rounded off to an 
 edge : polypiferous on both external and internal surfaces. 
 
 The structure of all the Ventriculidse is obviously designed for 
 the purpose of securing permanence of form, and thus safety and 
 free access of water to all the individual polyps. This object is here 
 effected by two means : first by the very remarkable structure of 
 the membrane already described ; secondly, by the regular, often 
 nearly hemispherical figure which the whole body assumes. 
 
 Specimens of Ventriculites are found of all sizes, and it can 
 easily be understood that, from the earliest period of their deve- 
 lopment, the same general form is assumed, which, as they increase 
 and spread, is still retained. Thus the question of their growth 
 seems to involve fewer difficulties than, at first sight, strike the 
 observer in respect to the other genera. 
 
 The access of sea-water was well secured in this genus, inas- 
 much as, though the folds of some species are close, the expan- 
 
 * I believe it to be always five ; but the difficulty which exists, from the 
 causes already named, in ascertaining these minute points in all individual 
 instances induces me thus to qualify the generality of this character. 
 
55 
 
 sion of the whole mass and relative size of the central cavity are 
 usually greater than in any other genus, thus compensating for 
 the frequently looser nature of the fold in the other genera. It 
 may perhaps be inferred that the ocean in which this genus 
 dwelt was subject to such modifications that its waters were less 
 liable to disturbance than those of the ocean in which Cepha- 
 lites and Brachiolites originated ; so that a larger size could be 
 attained in safety, and with less multiplied provision for secu- 
 rity against such disturbances. 
 
 The whole genus Ventriculites is characteristic of the upper 
 chalk. I have a few specimens which may possibly have come 
 from the middle chalk* ; but every such case is doubtful, while 
 I have carefully ascertained that the mass are certainly from the 
 upper chalk. 
 
 a. Simplices. 
 
 Inner and outer surfaces corresponding. 
 
 1. Ventriculites simplex. PL VIII. fig. If. 
 Membrane simple and without trace of fold : moveable processes 
 minute or absent. 
 
 This is the type of the whole family, and it is a most happy 
 circumstance that it exists. That existence at once destroys all 
 the theories as to the anatomy and physiology of this group of 
 fossils which have been suggested from examination of some spe- 
 cial forms only ; while it affords the means of demonstrating some 
 important points. It appears to be a rare species, as, though I 
 am fortunate in having myself obtained a very perfect series, I 
 have certainly never seen half a dozen specimens out of my own 
 cabinet. 
 
 When the underskin and polyp-skin are absent, as most often 
 happens, the specimen, if uninjured, is particularly beautiful. It 
 exhibits, over its whole surface, with the utmost regularity, that 
 square arrangement of fibre, with the subtending octahedra, already 
 described. When the underskin is present without the polyp- 
 skin the squares are not seen, but the whole surface is covered 
 with a much smaller-meshed tissue which appears dotted all 
 over, the dots being the polyp-cells. When the polyp-skin is 
 present it hides each of these appearances, and is marked only 
 
 * See Ann. and Mag. Nat. Hist. vol. xx. p. 337. 
 
 t It is not possible to give illustrations, in this work, of all specimens of 
 full size, several of them often exceeding in diameter the size of the page. 
 It will be understood, therefore, that all the specimens of the present genus 
 figured on PI. XIII., with the exception of fig. 9, are of specimens below the 
 usual size. 
 
56 
 
 with slight depressions in the places where the dots are seen 
 under the conditions last named. 
 
 The traces of the moveable processes appear slight, often quite 
 wanting, in this species. This is what might be expected if the 
 object of those processes was to sweep the surface clear of ob- 
 structions from the mouths of the cells. The surface being plain, 
 such obstructions would be far less liable to accumulate around it 
 than near those rugose surfaces which other species present, and 
 on the most rugose of which the processes are most strongly 
 marked. 
 
 I have specimens varying from 7 inches to 1 inch in dia- 
 meter. 
 
 It appears to me that the fossil called by Dr. Mantell Spongus 
 Townshendi (South Downs, p. 164) is nothing more than a V. 
 simplex, in flint. 
 
 2. Ventriculites impressus. PI. VIII. figs. 2 & 3*. 
 
 Membrane more or less slightly folded at upper part, but gene- 
 rally without regular figure : moveable processes frequently 
 conspicuous. 
 
 The basal portion of this species is usually plain like V. sim- 
 plex, but it soon exhibits marks of fold. These are sometimes 
 slight, sometimes deeper. They vary much in different speci- 
 mens, so that it might be thought advisable to distinguish varie- 
 ties. It has seemed to me, however, better to include all under 
 one name only. They may be generally known at once by the 
 basal portion being quite plain as in V. simplex ; by the much 
 less depth of the folds than in the next species ; by the depres- 
 sions being usually more or less oval instead of round ; and by 
 the want of any regular figure in the arrangement of the folds. 
 
 This being the first species in which the membrane is folded, 
 it will be well to consider the principle, if any, upon which the 
 fold in all the Ventriculidse is arranged ; and also to inquire whe- 
 ther the polyps existed, during the life of the animal, over the 
 inner parts of all the folds. 
 
 To treat of the latter question first : I have already intimated 
 that, as a general rule, exceptions may, as in Eschara and Flus- 
 tra, be found to this general rule, both external and internal 
 surfaces of the Ventriculidse were polypiferous. And notwith- 
 
 * The object of fig. 2 is to illustrate the very deceptive appearances which 
 the difference in the state of perfection of the fossils Mill occa^on. On the 
 lower part of that figure all remains of the actual body are lost. On the 
 upper part an Ostrea is seen, and, near it, a small part of the actual body 
 remains. The difference between this and the rest of the figure must be 
 very apparent. 
 
57 
 
 standing the closeness of the folds in many species, I believe that 
 the inner parts of all the folds were polypiferous also. The in- 
 quirer familiar with recent Polyzoa will find no difficulty in 
 realizing this as true, as he will be aware that these minute crea- 
 tures are often packed in spaces so close, that it would seem to be, 
 and perhaps is, impossible that all should be protruded at the same 
 time. This fact has attracted the attention of all observers. 
 Dr. Farre notices that the individuals of Halodactylus diaphanus 
 "are so closely set that there seems to be hardly room for 
 their several operations f and that, on this account, it is often 
 "scarcely possible to make any observations upon them" even 
 with the microscope*. And the most recent writer on the 
 Polyzoa, Sir J. G. Dalyell, in his ( Remarkable Animals of Scot- 
 land/ calls attention several times to the extreme complexity of 
 the mass and to the difficulty of even microscopic observation 
 on that account t- And in the plates of both these writers 
 (pi. 25. fig. 1, Farre ; pis. 43, 44 A, &c. Dalyell) the same fact is 
 well shown. It thus becomes obvious that, in recent allied forms, 
 individuals are packed in positions quite as close as, if not closer 
 than, in any of the Ventriculida?, and in positions apparently 
 more hazardous to the free action of the individuals themselves, 
 inasmuch as the form of the recent species specially referred to 
 is less fixed and unyielding than, from the very nature of their 
 structure, the Ventriculidse have been shown to have necessarily 
 been. 
 
 Having thus shown that there is no improbability that in fossil 
 forms the surface should have been thus closely covered, in all 
 its parts, by the polyps, I will add, that the results of positive 
 observation establish the fact that the inner parts of the folds in 
 the Ventriculidse were thus polypiferous. I have carefully dis- 
 sected several folds in specimens, both in chalk and flint, in 
 which the fold is the closest of any species, and I have found, by 
 aid of the microscope, the presence of the polyp-cells clearly and 
 unequivocally marked. 
 
 With respect to the arrangement of the fold, the elaborate dis- 
 section of many individuals of more than thirty different forms 
 of Ventriculidse has fully satisfied me that here, as elsewhere, 
 UNITY is prevalent ; and that the fold of the membrane forming 
 the wall of the pouch is not capricious and without method or 
 principle, as at first sight it might appear. I have fully satisfied 
 myself that every form and variety of foldj which different 
 species of Ventriculidse exhibit is based on a modification of the 
 
 * As before, p. 405. 
 
 t See, inter alia, pp. 233, 234, as to CelMarla loricnlata. 
 
 J Jt is necessary to distinguish the folding of the membrane forming the 
 
58 
 
 simple plait. Unimportant as the question may, at first sight, 
 appear, it is in reality of much importance. Every illustration of 
 the Law of Unity is, and ever must be, interesting to the student 
 of natural history, and important as the surest indication that 
 some point of truth has been attained. And without the guide 
 of some simple unity of this nature, it would be quite impossible 
 to make the structure of the different species of Ventriculidse 
 understood by the inquirer. The multiplied dissections which I 
 have made is a labour which few would have the inclination, and 
 as few the opportunity, to undergo. 
 
 In the present genus and section we shall meet with one spe- 
 cies in which the form of the simple plait is found without any 
 modification. In the present species, and in several others, the 
 primitive plait is not so obvious ; but the comparison of several 
 specimens, and especially, as is so often the case in the illustra- 
 tion of a Law of Morphology, of particular instances assuming a 
 somewhat abnormal character, will enable the careful observer 
 to trace, even in these species, this primitive plait. The limits 
 within which I am necessarily restricted prevent me from entering 
 more fully into, and illustrating at greater length, this question, 
 which I certainly regard as one of very much interest and im- 
 portance. In the present section of the genus Ventriculites, 
 in which every depression of one side answers to an entirely or 
 nearly corresponding depression on the other, the existence of 
 the plait is less material to the understanding of the character 
 of the wall of the pouch than where the fold assumes that com- 
 plexity which it does in the section Complicati. I shall dwell 
 only slightly upon the point, therefore, in describing individuals 
 of the present section, leaving the fuller illustration of it to those 
 specimens from which the illustrations of PL XIII. figs. 13, 14 and 
 15 are drawn, and which will I hope satisfy every careful and 
 candid inquirer that the most apparently differing external forms 
 may depend on the different modifications which the upper and 
 lower fold of a simple plaited membrane may undergo. 
 
 3. Ventriculites quincuncialis. PL VII. fig. 7, and 
 PL XIII. fig. 11. 
 
 Membrane deeply folded, usually from base to margin, of nearly 
 equal width the whole depth of the fold, and in regular quin- 
 
 wall of the pouch, where the whole body is a single pouch, from the subdi- 
 vision of the body itself into distinct lobes or branches as found in Bra- 
 chiolites. It will presently be seen that the wall of these lobes and branches 
 is marked by the same characters as the wall of the single-pouch forms now 
 under more immediate consideration. 
 
59 
 
 cuncial figure : moveable processes conspicuous : wall often 
 thick. 
 
 In this, as in the former species, must be included forms which, 
 at first sight, vary to a considerable extent, but which I conceive 
 to mark either distinct varieties or distinct ages. It is however 
 almost impossible, in specimens in a good state of preservation, 
 to mistake the species, the characteristic type of fold being found 
 in all with little variation. The wall is usually much thicker 
 than in V. impressus, and the fold begins from the base. In some 
 specimens the fold is so deep that a thickness of nearly two lines, 
 sometimes even more*, is attained in the wall of the pouch. 
 The folds are almost equally broad down their whole depth, so 
 that in this species the general aspect, both external and in- 
 ternal, is level, contrary to what takes place in V. muricatus, and 
 which difference will best be understood by comparing figs. 11 
 and 12 of PL XIII. which represent sections of these two species f. 
 
 It occasionally happens that the outer surface is not quite so 
 regularly marked as the inner, though still each depression on 
 one side is accompanied by a corresponding elevation on the 
 other. This variance is probably owing to some circumstance 
 connected with the process of the fossilization of the individual. 
 Hence even a partial angularity in the margin of the folds is, in 
 some such specimens, seen on the outer surface of the wall of 
 the pouch. Such specimens appear to have been partly crushed 
 before or during fossilization. The inside preserves all the usual 
 regularity of the fold. 
 
 It occasionally, though rarely, happens that the quincuncial 
 figure is not maintained ; in all specimens there must necessarily 
 be places where it is not truly kept, owing to the mode of increase 
 of the animal from base to margin. 
 
 The so-called Ocellaria, so often mentioned, belong to this 
 species. What are thus named are either young specimens, or 
 the lower parts of full-sized ones. These are usually nearly cy- 
 lindrical at the upper part and taper very gradually to the base ; 
 or form, altogether, a very acutely pointed inverted cone. 
 
 There is no species which affords more entire disproof of the 
 contractile theory than this. It is obviously impossible that any 
 membrane should contract in the way in which this form exists, 
 and with no variation whether forming a close cylinder or an 
 almost flat disc. 
 
 I am well-aware that this is the species to which appeal will 
 always be made in opposition to the views which have been stated 
 
 * I have an instance in which the thickness of the wall exceeds a quarter 
 of an inch. 
 
 t In the specimen figured on PI. VII. fig. 7. the folds are unusually 
 narrow, but the quincuncial character is well-displayed. 
 
60 
 
 as to the fold of the membrane. It will often be triumphantly 
 referred to as affording evidence of perforation. There may be no 
 primd facie reason why the Ventriculidse should not, like the re- 
 cent Retepora, have been perforated. When however it is found, 
 through a vast number of differing forms, that one plan, that of 
 the folded membrane, has been unquestionably adopted; and 
 when it is found that in no other case in which Ventriculitic 
 structure is present is a thickness of more than five of the solid 
 squares attained by any part of the fold which forms the wall of 
 the pouch, while it is found that, in this species, the thickness of 
 the entire wall varies from one line to three lines, thus exhibit- 
 ing, if the actual wall be but a perforated mass, the extraordinary 
 and, upon every principle, anomalous fact of there being no con- 
 stancy* in the structure of the central polypidom ; there is cer- 
 tainly every reason to conclude that Unity is not violated in this 
 case, and that, like others of the family, this is truly a folded and 
 not a perforated membrane. The various deceptive appearances 
 which the mere fossil may put on have been already more than 
 once remarked. In the examination of the present species, 
 above all, the greatest skill and care are necessary, because it is 
 obvious that the very depth and narrowness of the fold must, if 
 any abrasion or injury of the external surfacef take place, de- 
 stroy the traces of the membrane covering the top of each 
 depression, and because the small size of the depressions and 
 elevations renders it very difficult to follow the fold which forms 
 them from one surface to the other. Feeling therefore to the 
 fullest extent the difficulty of ascertaining the actual fact in this 
 case, and at the same time the importance of ascertaining that 
 fact, both as an exception, if it should prove such, to the Law of 
 Unity, and as a determination of a point in paleontology which 
 has been now mooted for more than thirty years, I carefully dis- 
 sected numerous specimens, both in flint and in chalk, and both 
 with the slitting- wheel and the needle, and followed up and down 
 the so-called (sometimes) cells or (sometimes) perforations to their 
 terminations. The clear and unequivocal result has been that the 
 so-called Ocellaria is no anomaly : that the Law of Unity has not 
 its exception here ; that this is a true case of fold of membrane : 
 finally, that I have at this moment before me specimens in which 
 
 * It will be obvious to the careful inquirer that there is nothing anoma- 
 lous in the varying depth of the folds, inasmuch as, in all species, that depth 
 will naturally be dependent, more or less, on the age of the individual. 
 
 f In every instance, both in chalk and flint, with very rare exceptions, 
 some of the body adheres to the matrix. In the present species, where the 
 bases of the folds are so small, it will necessarily, therefore, often happen 
 that just those bases adhere to the matrix. Hence alone will result an ap- 
 parent perforation of the wall, the remains of the membrane not being pre- 
 sent, and not therefore traceable, over those bases. See PI. XIV, fig. 14. 
 
61 
 
 I can as clearly see the fold in and out of that membrane as it is 
 seen in fig. 1 1 . PL XIII. I wish to be specific and precise on this 
 point because I know that a bare description will be questioned. 
 1 am obliged to state that such a reality as is represented " mag- 
 nified" on p. 279. fig. 2 b of Dr. Mantell's ' Medals of Creation' 
 never had existence * in the nature of the animal of which fig. 2 
 gives a view of the natural size of the fossilized remains. 
 
 In the museum of the Yorkshire Philosophical Institution 
 there is a fragment of a specimen of this species almost discoidal, 
 which still measures more than nine inches across, and, when 
 entire, must have been of still greater size. This is an unusually 
 large size for the species to attain, but the variation in that re- 
 spect is very great. There is another interesting specimen of 
 this species in that museum in which the whole body has become 
 converted into chalcedony. The outer surface is nearly entire, as 
 is also a considerable part of the inside. The structure is, as 
 usual in solidified chalcedonic specimens f, obliterated throughout 
 the greater part, but it can, in places, be still clearly traced. The 
 individual underwent pressure before or during the process of 
 fossilization, and also some decomposition, so that the external 
 character has become somewhat modified, and it is only in parts 
 that the internal arrangement can, at places of section, be seen. 
 The peculiar state of the specimen, solidly chalcedonic without 
 the presence of any true flint, renders it however very in- 
 teresting J. 
 
 4. Ventriculites muricatus. PL XIII. figs. 1 & 12. 
 Membrane deeply folded, without regular figure : folds unequal 
 in width throughout their depth: moveable processes often 
 conspicuous : wall of moderate thickness. 
 
 The point wherein this species differs most essentially from 
 the last will be at once understood by comparing figs. II & 12 
 of PI. XIII. It has been already explained that, through the uni- 
 formity in the breadth of the fold, the general surface of V. quin- 
 
 * And see p. 280 : the polyparium of this species is no more " calca- 
 reous " than that of any other of the Ventriculidae. Treated with acid it is 
 acted on in precisely the same way and to the same extent as are all speci- 
 mens of Venlriculid'ae preserved in flint, and leaves an exquisitely delicate 
 cast of the body of the animal and of the structure of its polypidom. 
 
 f See before, p. 13. 
 
 j It has already been stated (p. 14), that " in general when any part 
 of the soft substance of a body encased in chalk decomposed, its place was 
 soon rilled up with particles of chalk." The above instance is one of the 
 rare exceptions to this general fact; and the peculiar condition of the 
 Yorkshire chalk (which is much more compact than that of Kent, &c.) may 
 lead us to expect more numerous instances of these exceptions from that 
 region than from the chalk of the south-east of England. 
 
62 
 
 cuncialis is smooth. It will be perceived that where, as in fig. 12, 
 the folds are broad on each surface at their margin and narrow 
 to a point at their base, the effect must be to give to the whole 
 wall of the pouch a rough surface. The wall is usually not very 
 thick. I have never seen it attain so great a thickness as V. 
 quincuncialis often does. The height and depth of the folds 
 render it difficult ever to confound it with V. impressus, in which 
 also the same smooth general appearance of the surface as in 
 V. quincuncialis is usually maintained, and for the same reason 
 as in that species. 
 
 5. Ventriculites tessellatus. PL XIII. figs. 2, 3, 4. 
 
 Membrane folded in regular quadrilateral and rectangular figures 
 usually more or less oblong : wall of moderate thickness. 
 
 In Mr. Morris's Catalogue mention is made of a Ventriculites 
 quadratus, and reference is made to Goldfuss, pi. 33. fig. 1. I 
 have already* stated that this is no Ventriculite. The description 
 is, " seriebus pororum oblongorum rectis parallelis decussantibus } )} 
 an error into which I cannot conceive it possible that Goldfuss 
 could have fallen if he really had the present species before him. 
 And the magnified sketch given in fig. 1 b of the same plate, 
 unless fancy and not actual observation be there copied, differs 
 as widely from any characters of the present species, and from 
 any form of Ventriculitic structure, as does the description which 
 has been cited. 
 
 This form is rare ; and it is still rarer to find a perfect speci- 
 men. I have several fragments, excellently well displaying struc- 
 ture and fold, and during the summer of 1847 I was fortunate 
 enough to find a perfect specimen, with roots and margin, and 
 both body and cast, entirely perfect, and five inches in diameter ; 
 but I was so unfortunate as for it to break in pieces before I could 
 convey it home, a danger to which the collector of these delicate 
 fossils is peculiarly liable. Though this loss is not easily to be 
 replaced, the opportunity of verifying the entire form of the spe- 
 cies was important. The rarity of the species has prevented me 
 from making sufficiently extended observations on the moveable 
 processes to state, with confidence, anything definite in regard to 
 them. 
 
 In this species the plaits may often readily be traced, with the 
 oblong depressions running along them on each face of the wall. 
 This being borne in mind, the figures often displayed on section 
 will readily be understood, when, instead of the regular alter- 
 nation of depressions seen at one place of section, as in fig. 4 
 
 * Ante, p. 6. 
 
63 
 
 (PL XIII.), the character represented in fig. 3 will, at another 
 place of section, be seen. 
 
 6. Ventriculites cavatus. PL XIII. fig, 5. 
 
 Memhrane folded in concavities of considerable breadth, rarely 
 
 regular in disposition : wall of moderate thickness : moveable 
 
 processes often well-seen. 
 
 This species can never be confounded with the last. The shape 
 of the fold, its sloping edges, and the want, in general, of regular 
 figure, at once distinguish it. A linear arrangement of the de- 
 pressions is often traceable, marking the places of the primitive 
 plait. 
 
 This species might readily be divided into two or three varie- 
 ties, marked by the fewness or abundance, the small or large 
 size, and the greater or less degree of regularity, of the folds. 
 The principle of the fold, which in all cases is the ground- 
 work of the present classification, appears however to be the 
 same in all these cases, and they may, not improbably, be but 
 modifications indicative of a difference of age. The concavities 
 are usually somewhat elongated in specimens of which the pouch 
 is closely cylindrical. 
 
 In this case, as in all others, it is necessary that both sides, or 
 a good section, of the specimen should be examined, as V. radiatus 
 will be found to be marked on the inner surface by characters 
 bearing some superficial resemblance to those distinguishing both 
 surfaces of the present species, while V. bicomplicatus bears, both 
 on the external and internal surfaces, characters which may some- 
 times mislead the observer. A section of the present species how- 
 ever will show the wall much thinner and the cavities shallower 
 than in either of those species, while, in well-preserved specimens, 
 the entirely different character of the fold cannot fail to be seen. 
 
 7. Ventriculites striatus. PL XIII. figs. 6 & 13. 
 Membrane folded from base to margin in regular close and sim- 
 ple plaits : processes very conspicuous : wall of moderate thick- 
 ness. 
 
 This is an exceedingly interesting and very marked species. 
 The plaits, which have heretofore been traceable only by careful 
 comparison and an effort of induction, become in this species the 
 one characteristic feature. Not a trace is present, on either sur- 
 face, of the " tubuli" assumed to be polyp-cells or absorbents 
 of Dr. Mantell, and the presence of which is expressly stated by 
 him to be characteristic of the genus and species* ; upon the as- 
 sumption of which indeed all his views and descriptions have 
 
 * See ante, p. 5. 
 
64 
 
 been based. The inquirer will be well-prepared for this entire 
 absence of any such tubuli from its having been seen that, in the 
 preceding species, the depressions (which alone are what could, 
 even on a superficial observation, be called tubuli) have the 
 greatest diversity of form and character, a diversity which at 
 once negatives their being of the nature assigned to them by 
 Dr. Mantell. It has moreover been seen that, as a matter of fact, 
 they are in no instance tubuli, but that they can, on the contrary, 
 be traced as folds of the membrane forming the polypidom itself. 
 
 On the other hand, the present is the only species which could 
 seem to lend any support to the contractile theory. The observer 
 however who has studied the several preceding species will pro- 
 bably ask for no proof that the present does not stand out as an 
 anomaly and an exception from them all. If he need such, he 
 may be referred to the observations already made *, generally, on 
 the matter. These have full application in this as in every other 
 case, and sufficiently show that this species offers no violation to 
 that Unity which prevails through every branch of the present 
 inquiry. 
 
 Different specimens vary in the size and depth of the plait. 
 It is suggestive, perhaps, as before, of difference of age. It is 
 where the plait is deepest and broadest that the moveable pro- 
 cesses become most conspicuous. 
 
 The plaits are not narrow at the base and of increasing size 
 as they approach the margin. They maintain the same size from 
 base to margin, and the increase of surface is effected by the 
 increase in number of plaits as the margin is approached ; an in- 
 crease effected by the division, from time to time, of one plait 
 into two, each of equal size with the original one, a mode of 
 increase which generally prevails in all species in which the plait 
 can be distinctly followed. The species is not common. 
 
 b. Complicati. 
 Inner and outer surfaces not corresponding. 
 
 1. Ventriculites mammillaris. PI. XIII. figs. 7 & 14. 
 
 Inner plaits simple and regular : outer plaits raised in large hol- 
 low bosses at regular intervals : processes very conspicuous : 
 thickness of wall considerable. 
 
 This is the first instance in which we find the direction of the 
 fold changed between the external and internal surfaces. There 
 is no species more strongly marked, and none which affords a 
 better illustration of the value and importance of a principle of 
 Unity as the essential and most valuable means to true scientific 
 
 * Ante, p. 17. 
 
65 
 
 research. The fact of the mammillated external appearance of 
 this species is easily seen and as easily recorded : so is also the 
 fact of the striated internal appearance. But it would be difficult, 
 by any description, to make it clearly understood how such ap- 
 parently contradictory appearances could result from the folding 
 of a simple membrane*, until it had been ascertained, by careful 
 and multiplied dissections, that this and all other characters of 
 the fold of the membrane forming the wall of the pouch are 
 grounded upon one single and simple unity of plan of which all 
 present clear and intelligible modifications. 
 
 On the outside of this species as usually seen in a fossil state, 
 and as it must usually have appeared in a recent state, no mark 
 of the plait is seen. Large rounded elevations scattered, at first 
 sight irregularly, over the whole surface, are all that meet the 
 eye. If however the reader will carefully consider figs. 13 and 
 14 of PI. XIII. he will see, in fig. 13, the simple plaits, regular and 
 uninterrupted, within and without, as in V. striatus-, while in 
 fig. 14 he will see that the perfect plait is still present, and that 
 the inner surface [the lower part of the figure] is still simple and 
 uninterrupted, but that the outer plaits are interrupted along 
 their whole length by rounded elevations, not solid, but hollow, 
 elevations of the membrane into those shapes instead of the 
 plait being plain and simple. These figures are somewhat ex- 
 aggerated in size in order that the principle may be more clearly 
 understood. 
 
 It will be found that in each of the forms which follow, and 
 
 * As the extraordinary complications in the fold of the membrane of 
 many species were gradually developed by multiplied dissections, 1 long de- 
 spaired of being able, by any descriptions, to make them understood. When 
 by degrees the unity of plan which I have endeavoured to indicate above, 
 as the groundwork of all that complexity, opened upon me, t felt an im- 
 portant key to have been obtained, which experiment proved to be appli- 
 cable to every case. I indulge some hope that the development of this 
 unity may be of a utility beyond the mere understanding of the forms now 
 under discussion however interesting those may be to myself and Others. 
 In the descriptions of numerous tissues in human and other branches of 
 anatomy, I have often myself felt the want of some clear and simple basis 
 of unity in the descriptions attempted of those tissues ; which want has 
 caused them to be often unintelligible. Any pains which the investigation 
 of the present subject may have cost will be more than rewarded if the sug- 
 gestion of an unfailing unity in the arrangement of all complex tissues shall 
 be felt to be (as I cannot doubt that it is) generally applicable. It seems to 
 me, that as, by application of this principle, forms varying so entirely as it 
 will be found that many of the Ventriculidae do in mere general external cha- 
 racters, and therefore heretofore classed in entirely different natural groups, 
 are now demonstrated to belong to one group, so the application of the like 
 principle may, in many points of anatomy, lead to the discovery of intimate 
 relations, not now suspected, between tissues which appear very different 
 in structure. 
 
66 
 
 which appear so widely different from each other in all that meets 
 the eye of the general observer, the characters which each exhibits 
 may be as readily understood as in the present species by taking 
 as the basis that simple Unity of plan which has been thus illus- 
 trated. 
 
 2. Ventriculites latiplicatus. 
 Plaits very broad and deep : outer plaits simple and regular : 
 
 inner plaits inclining, at intervals, towards, and anastomosing, 
 
 a little below the surface, with, adjoining plaits : wall very 
 
 thick. 
 
 This is a very interesting species. It presents us with the first 
 instance of the plaits departing from their entire individuality, 
 and of a connexion being formed between adjoining ones. In the 
 present species that connexion is merely by a slight anastomosis 
 at distant intervals. There being no depression in the plait, on 
 either surface, the access of water is perfectly preserved without 
 any additional provision such as we shall find in V. radiatus 
 and many others. 
 
 In V. mammillaris the departure from the simple plait took 
 place on the outside. In the 
 present species it takes place on 
 the inside, and is more marked 
 in its character and degree ; but 
 it is none the less an illustration 
 of the principle of Unity which 
 it has been endeavoured to ex- 
 plain, being clearly but a modi- 
 fication of the simple plait. 
 This will be understood from 
 the following figure, in which 
 b, b, b are the lower plaits; c, c, c y 
 the upper plaits; and a t a, a, 
 the points of anastomosis. We shall, in the next species, find 
 the complexity of the fold much increased. 
 
 The wall of this species is thicker than that of any other spe- 
 cies of the genus Ventriculites. 
 
 3. Ventriculites decurrens. PI. XIII. fig. 8. 
 Plaits often irregular in direction : outer plaits constricted at 
 distant and unequal intervals : inner plaits depressed at short 
 and nearly equal intervals ; bulging on each side around de- 
 pressions till adjoining plaits meet and open into each other : 
 processes conspicuous : wall of moderate thickness. 
 
 Var. tenuiplicatus. PL XIII. fig. 9. 
 Plaits close and delicate : outer plaits nearly regular in direction 
 
67 
 
 from base to margin with slight occasional constrictions : 
 inner plaits depressed at regular intervals ; bulging on each side 
 around depressions till adjoining plaits meet and open into 
 each other : processes conspicuous : pouch very often nearly 
 cylindrical : wall thin. 
 
 Marked specimens of these two varieties are very distinguish- 
 able, but, upon the principles to which I have confined myself in 
 discriminating species, I do not think it proper to separate them ; 
 the elements of the modification of the fold being similar in each 
 variety. Specimens in which the two varieties run into each 
 other are not unfrequently found. 
 
 The variety tenuiplicatus is a most beautiful and delicate fossil. 
 I have several specimens on which the polyp-cells are finely seen. 
 
 In examining each of these varieties the importance of the clue 
 afforded by the principle already explained of Unity in the fold 
 of the membrane becomes strongly felt. Let the reader compare 
 figs. 13, 14 and 15 of PI. XIII., always remembering that fig. 13 
 is the basis of every modification. In fig. 15 he will see the lower 
 plait [it is immaterial for the present purpose whether that is to 
 be looked at as the inner or outer surface] retaining its simpli- 
 city, while along the top of the upper plait he will see a series of 
 circular depressions at short and equal intervals. At the sides of 
 these depressions the membrane bulges out, in order to pre- 
 serve the walls of the plait below the depression clear from con- 
 tact*, until the bulgings on the two adjoining plaits meet and 
 anastomose. These then open into each other in the centre of 
 the place of anastomosis, leaving, however, below them and on 
 each side a clear space ; so that strength and security are gained 
 to the whole mass, and, consequently, the safety of the individual 
 
 * This is but one other among the numberless beautiful illustrations 
 afforded by the anatomy of the Ventriculidae of obvious design and adap- 
 tation. Did these walls touch, there would be a great loss of surface and 
 little or no additional strength. By the actual arrangement there is a vast 
 gain of both surface and strength, as well as an additional security for free 
 access of sea-water to all parts of the inner surface. It has already been 
 seen that this additional provision not being needed in F. latlplicatm is not 
 found in that species. It is proper to add, that the fact that, around the de- 
 pressions, the adjoining plaits do not only anastomose but open into each 
 other has been ascertained by most careful examination and dissection of 
 specimens in chalk and in flint ; but that, having clearly ascertained the fact 
 in instances where there is thus such an obvious purpose for such an arrange- 
 ment, I cannot doubt that, under similar circumstances of structure in ad- 
 joining parts, such is always the arrangement of the fold in this particular, 
 though the minuteness of the parts generally prevents the possibility of 
 clearly distinguishing it in chalk specimens. The place of the opening does 
 not usually exceed, in the necessarily somewhat collapsed state in which 
 it is exhibited in the fossil condition, the 40th of an inch in diameter, a 
 space which the iron stain in the chalk is more than sufficient to obliterate. 
 
 E2 
 
68 
 
 polyps is ensured, together with a great increase of polypiferous 
 surface. 
 
 I am aware of no instance among the Ventriculidse in which 
 a mere ridge or flap of membrane is added on, as this 
 would practically be if the membrane forming the sides of these 
 depressions came in contact with that forming the sides of the 
 plaits. There are, on the other hand, very many cases in which, 
 as in the present species, additional strength to the whole mass 
 is afforded by a means which gives a very great additional surface 
 for the development and security of that life, the manifestation 
 and multiplied means of enjoyment of which were the end of the 
 existence of the whole creature. 
 
 On each side of the places where these meeting bulgings blend, 
 the vacant spaces between the plaits put on, naturally, from the 
 shortness and regularity of the distance between the bulgings, a 
 circular form. As one of these must, of course, lie between each 
 double pair of the depressions, a quincuncial figure is thus as- 
 sumed by the whole (see the darker spots in fig. 15 of PL XIII.). 
 The inquirer will thus at once understand the appearance pre- 
 sented by the inner surface of many species of Ventriculidse. He 
 will clearly see that the " tubuli " of Dr. Mantell are things with- 
 out existence. He will understand that though, at the surface of 
 a fossil of which all the interstices are filled with chalk, all the 
 circular marks appear alike, yet the dissection of alternate rows 
 of them will reveal very different conditions ; one row consisting 
 of more or less shallow (in the present species usually rather 
 shallow) and closed depressions; the other having no true de- 
 pressions at all, but consisting of cavities extending to the bottom 
 of the intermediate plait, and indeed, before and behind, under- 
 neath the intermediate bulgings, running into and forming part 
 of the longitudinal cavity of that plait. 
 
 The wall of the variety tenuiplicatus being usually thin, the de- 
 pressions on the inner plaits, though shallow, are sometimes to 
 be seen through on the outside, if the specimen has come very 
 clean out of the chalk. It is very rarely, however, that specimens 
 can be got out of the chalk in this clean manner, portions of the 
 matrix usually filling up the spaces between the plaits. In the 
 recent state, and when, instead of being in the collapsed condi- 
 tion in which we find the fossils, the animals were alive and fully 
 distended with all- their fluids*, no doubt the lower surface of 
 all the depressions on the top of the inner plaits could be seen 
 from the outside. The living creature must have appeared as 
 composed of a number of plaits which, on their inner surfaces, 
 
 * Every reader must perceive the difference between the comparative 
 states of coflapsion and distension and those of contraction and expansion. 
 
69 
 
 exhibited numerous shallow depressions ; and which plaits, on 
 each side of these depressions, were united together by small 
 transverse tubules, the mouths of which tubules were directed 
 towards the outer surface of the whole body, and opened, always, 
 into the interspaces between the outer plaits. 
 
 It will be obvious that sections, whether made longitudinally 
 or transversely to the plaits, will vary in numberless ways in the 
 figure which will be seen, according to the particular point of 
 the plait through which the section passes. This must be re- 
 membered when the section seen at the edge of a flint or frac- 
 tured chalk specimen is examined. This observation applies as 
 well to the present species as to almost every other which will 
 subsequently come under notice. 
 
 The plaits of V. decurrens are not very regular in their course. 
 On the contrary, the outer plait, where the course is easily traced, 
 is often very winding, though several generally run parallel to 
 each other. The constrictions occur at irregular and distant in- 
 tervals, and cause the surface to appear as if several of the round 
 elevations of V. mammillaris had run together down the surface, 
 a modification which may truly be considered to represent the 
 fact. Hence the specific name. The constrictions do not extend 
 the whole depth of the plait. They do not appear ever to ex- 
 tend so deep as to interfere with the mode of fold characteristic 
 of the inner surface. It is owing to the winding course of the 
 plaits in this species that the rings seen on the inside do not 
 assume, as in V. radiatus, a perfectly regular quincuncial figure ; 
 the places of the depressions and accompanying characters of 
 structure varying, of course, in places, as the plait winds. In 
 V. tenuiplicatus, where the plaits are more regular, the quincun- 
 cial figure on the inside is correspondingly more regular. The 
 constrictions on the outer plaits of that variety are much slighter 
 than in V. decurrens. They resemble much more a very slight 
 exhibition of the normal characters of V. mammillaris. They thus 
 rarely interfere with the continuous course of the plait. 
 
 4. Ventriculites radiatus. PI. XIII. figs. 10, 15. 
 
 Plaits broad and deep : outer plaits regular, and with an occa- 
 sional lateral connection : inner plaits deeply depressed at 
 short and equal intervals ; bulging on each side around de- 
 pressions till the adjoining plaits meet and open into each 
 other : processes very conspicuous : wall thick. 
 I have retained for this interesting species the name radiatus 
 of Mantell, as being the species which comes nearest in super- 
 ficial external and internal appearance to his descriptions already 
 cited*. 
 
 * Ante, p. 4. 
 
70 
 
 This species unites the characters observed in V. latiplicatus 
 and V. decurrens. In the former mere points of connection exist, 
 on one surface, between the plaits ; in the latter the remarkable 
 depressions, bulgings and openings into each other last described 
 exist, ako on one surface only. In the present species the latter 
 characters are found on the inside, the depressions being, how- 
 ever, deeper than in V. decurrens ; while, the plaits being deep, 
 they exhibit, on the outside, and at distant intervals, points of 
 mere connection (no depressions being present on that surface) 
 as in V. latiplicatus. 
 
 The difference between the constrictions which mark the 
 outer plaits of V. decurrens, and the points of connection, with- 
 out any constriction, which exist between the outer plaits of the 
 present species, is alone sufficient to distinguish the two. In 
 addition to this, however, the present species is much thicker 
 and more massive than V. decurrens ; it often attains a thickness 
 in the wall of the pouch nearly as great as V. latiplicatus. The 
 regularity in the direction of the plaits is another distinguishing 
 character. 
 
 Owing to the regularity of the plaits in this species, the figure 
 assumed by the inner folds is very regularly quincuncial. 
 
 The inquirer must be careful not to be misled by inspection 
 of an inner surface only into determination of this species, as 
 such surface in either V. cavatus or V. bicomplicatus, especially 
 the latter, and sometimes in V. decurrens, will be difficult to di- 
 stinguish. It cannot be too often insisted on that parts of both 
 surfaces, as well as sections, should be examined before determi- 
 ning species with certainty. 
 
 This species is very local in its distribution, being very rare in 
 many places. 
 
 5. Ventriculites bicomplicatus. 
 
 Plaits broad and deep : both outer and inner plaits deeply de- 
 pressed at short and equal intervals, which alternate in ad- 
 joining inner and outer plaits ; bulging on each side around 
 depressions till adjoining plaits open into each other: pro- 
 cesses very conspicuous : wall very thick. 
 
 In the present species we meet 
 with a much greater complexity of 
 fold than in any preceding one. 
 The plaits on both surfaces un- 
 dergo a marked modification which 
 resembles the modification seen on 
 the inner plaits of V. radiatus. The 
 form of the depressions on the 
 external surface are, however, ge- 
 nerally oval instead of round, as in fig. E. 
 
71 
 
 The descriptions which have been given of the mode of fold 
 on the inner surface of V. decurrens and V. radiatus render it 
 unnecessary to dwell on the nature of the modifications which 
 the present species exhibits, and which are but the reduplication 
 of the remarkable characters exhibited in the species just named. 
 It may be suggested, however, that sections of this species will 
 vary so much, according to the direction in which they are taken, 
 that no one figure can be depended on as certainly indicative of 
 the species, without very extended and careful comparison. Not- 
 withstanding, however, the complexity of the folding in the pre- 
 sent species, the application of the observations already made in 
 respect of V. decurrens will make.it clear that the free access of 
 sea-water would be perfectly maintained to all parts of the very 
 extended polypiferous surface which is thus gained. 
 
 In all these deeply folded species we have seen that the move- 
 able processes are very conspicuous. They are thus conspicuous 
 on that portion of the fold which is most exposed. There would 
 obviously not be room for their operation within the folds, and 
 the traces of them are not found there as they are on the external 
 surface. If their object be that which has been suggested*, their 
 presence and action on this exposed external surface would be 
 amply sufficient for the protection of the polyps in the deeper 
 recesses of the folds, and the presence of the processes there 
 would be useless. The absence of their marked development in 
 those parts is therefore in harmony with that admirable adapta- 
 tion of means to ends which the results of every part of the pre- 
 sent investigation have displayed. 
 
 In examining the various and complicated forms which have 
 been thus noticed as included within the genus Ventriculites, 
 nothing is more striking than to find that, be the wall of the 
 pouch thick or thin, the thickness of the membrane itself remains 
 always the same. It may be folded up in the most complicated 
 way, but it still retains, in any single piece of it, precisely the 
 same characters as have been described as typical in respect to 
 V. simplex. Were other facts wanting, this would alone go 
 far to induce the conclusion which other courses of investigation 
 have already seemed to demonstrate, viz. that we have before us 
 a true Polyzoic polypidom ; and that polypidom one of the most 
 admirable construction and contrivance. No less interesting 
 illustrations of that conclusion will be found in the widely differ- 
 ent forms and modifications of folding which remain to be exa- 
 mined. 
 
 * Ante, pp. 34 and 56. 
 
72 
 
 Genus CEPHALITES. 
 
 Character. Pouch-shaped : very constant in size and dilatation : 
 cavity usually regular and with a single opening ; sometimes 
 winding and with more openings than one : membrane form- 
 ing the wall of the cavity always deeply folded : marginal 
 edges and, sometimes, most prominent points of the plaits 
 attached to a simple apolypiferous membrane stretched across 
 their whole breadth and forming the upper margin or head of 
 the wall : membrane of wall polypiferous on both external and 
 internal surfaces. 
 
 The differences between the genera Cephalites and Ventriculites 
 are so broadly marked that, except in one or two species, it would 
 be difficult to confound even fragments of the two. In every 
 species of Cephalites the head is conspicuous and unmistakeable. 
 This very remarkable peculiarity is alone sufficient to distinguish 
 the genus*. 
 
 The provisions found through the whole family of Ventricu- 
 lidse for ensuring the free access of sea-water to all parts of the 
 surface, and for securing permanence of form as one great means 
 to that end, have been already noticedf. The present genus 
 offers fresh and most remarkable illustrations of those provi- 
 sions. 
 
 In every species of this genus the fold is, comparatively to the 
 size of the whole body, much deeper and broader in many 
 species positively much deeper and broader than in any species 
 of the genus Ventriculites. The size also is much smaller than 
 the average size of the Ventriculites ; the height of specimens 
 of the present genus seldom exceeding two inches, rarely attain- 
 ing three inches J. The form is never expanded, as usual in 
 Ventriculites, but, with few exceptions, approaches nearly to the 
 cylindrical, as in V. tenuiplicatus. 
 
 Extent of surface was thus gained in this genus by the increased 
 depth and complexity of the fold. But this depth and complexity 
 would endanger the safety of the polypiferous surface were there 
 no special provision for maintaining the normal position of the 
 individual plaits. This was perfectly effected, and at the same 
 
 * See ante, p. 51. 
 
 f Ante, pp. 46, 54. It was the circumstance of the Ventriculidae being 
 polypiferous on both surfaces that rendered these provisions so necessary. 
 In Halodactylus, &c. one surface only is polypiferous. See note J p. 46. 
 
 J Hence all the figures of this genus are of specimens of average size. I 
 have much pleasure in acknowledging here the pains and care bestowed by 
 Mr. Sowerby over these plates. The novelty of the forms and structure 
 presented many difficulties, especially as the engravings were made only 
 from my drawings. But nothing can be more generally successful or truth- 
 ful than the figures which Mr. Sowerby has realised. 
 
73 
 
 time with great simplicity and beauty, by stretching across the 
 flat upper edges, or, in a few cases, the more prominent points*, 
 of the plaits a simple and entire membrane t, which, spread over 
 the whole breadth of those edges and from point to point of those 
 prominences, retained all the plaits securely in their position ; 
 thus ensuring the safety of the whole colony and of the entire 
 polypidom which was covered by it. See PL XIV. 
 
 The general constancy in the size and form of specimens of 
 this genus throws difficulties in the way of the question of growth. 
 It is not easy to understand why we do not find young indivi- 
 duals of this genus as of Ventriculites. It has occurred to me 
 that, probably, the ocean in which this genus dwelt being, appa- 
 rently, a more disturbed one than that in which the Ventriculites 
 dwelt J, and the head possibly not forming till a certain age and 
 size had been attained, individuals dead or destroyed below that 
 age very rapidly lost their form and are therefore found only as 
 shapeless masses. I do not suggest this solution of the difficulty, 
 however, without considerable hesitation. 
 
 The whole genus Cephalites is characteristic of the Middle 
 Chalk. I have never found a single specimen which I could with 
 any probability refer to the Upper Chalk, though it may be ex- 
 pected that some forms will be found which endured into that 
 later epoch. Certainly none have been ever yet found in the 
 Lower Chalk. 
 
 a. Annulati . 
 Head narrow and flat : plaits compact and regular. 
 
 1. Cephalites longitudinalis. PL VII. fig. 1, & PI. XIV. fig. 1. 
 
 Plaits delicate but often deep : outer plaits slightly winding : 
 inner plaits depressed at short and regular intervals ; bulging 
 on each side around depressions till the adjoining plaits meet 
 and open into each other : processes very conspicuous : wall 
 moderately thick. 
 
 This species much resembles in external aspect the smaller 
 cylindrical specimens of Ventriculites tenuiplicatus. It is however 
 smaller than that species usually is, the plaits less winding, and 
 
 * These latter cases form, however, no exception to the principle of the 
 marginal edge of the plaits being always attached to the cephalic membrane. 
 The cases in which prominent points of the plaits are attached to the head 
 are cases of an additional provision for security, In those cases, as in all 
 others, the marginal edge of the membrane, after having undergone all its 
 varied modifications of fold, reaches and is attached to the head. See the 
 description of C. campanidatus and C. constrictus. 
 
 f As to structure and nature of this see ante, pp. 24, 37. 
 
 J See ante, p. 55. See ante, p. 52. 
 
74 
 
 the wall thicker. The depressions on the inside also are generally 
 smaller, closer, and more regular than in that species. The head 
 alone is sufficient to distinguish the two at a glance. 
 
 This is the only species of Cephalites in which the longitudinal 
 fold remains unmodified on the outer face. Hence its specific 
 name. A transverse section of it is seen on fig. 1 of PL VII. 
 
 It is a rare and delicate species : indeed all the species of the 
 present genus are rare. They do not seem to have abounded in 
 the older seas of the Middle Chalk as the Ventriculites did in the 
 Upper Chalk. Though thus rare, however, their modifications are 
 not the less clearly marked. 
 
 In regard to the head it is proper to remark, that while, 
 throughout the present division of this genus, its breadth will 
 always be found a very near approximation to that of a transverse 
 section of the plaits, there is a slight variation in this respect in 
 individual specimens. The head often slopes a little outwards, so 
 
 that a section of the entire body presents an 
 outline as in fig. F, in which a b is the sec- 
 tion of the head. The outline of the head 
 is always quite as sharp and well-defined as 
 in this figure. The relative arrangement 
 and proportions of the head and the plaits 
 are such that specimens of this division can 
 never be confounded with any belonging to 
 the section Dilatati. It is very rarely in the 
 present division that there is any rounding, 
 or departure from the nearly flat character of 
 the head; a character, on the other hand, 
 never present in the Dilatati. 
 
 It is proper to notice that, in every species of this genus, in 
 order to give full strength to the head, the depressions, bulgings, 
 and other modifications of the fold, where it does not rise, as 
 in C. campanulatus, in a simple form, are so arranged that the 
 membrane of the inner wall, where it adjoins the head, is always, 
 and that of the outer wall most frequently, expanded by a lateral 
 bulging of the plait, so as for the adjoining plaits to meet just 
 at the point of union of the wall with the head. Thus the 
 whole of the inner, and often of the outer, edge of the head is 
 continuously attached to the wall, an arrangement of much im- 
 portance. On this inner edge the membrane often rises up in a 
 narrow and slightly prominent ridge above the otherwise smooth 
 surface of the head. 
 
 2. Cephalites guttatus. PL XIV. fig. 2. 
 
 Plaits broad and deep : outer plaits raised in large hollow bosses, 
 often elongated ; adjoining plaits having an occasional lateral 
 
75 
 
 connection : inner plaits depressed at regular intervals ; bulging 
 on each side around depressions till adjoining plaits meet and 
 open into each other : processes very conspicuous : wall usu- 
 ally thick. 
 
 Nothing can better express the usual character of this species 
 than the term guttatus. The outer surface looks exactly as if 
 sprinkled with drops of a viscid fluid which had just begun to 
 run together, in some instances to a greater, in others to a less 
 extent. It is thus generally well distinguishable, even on the 
 outside, from Ventriculites mammillaris. The plaits being much 
 broader than in C. longitudinalis , the depressions on the inner 
 plaits are larger than in that species. 
 
 The lateral connection between adjoining external plaits, as in 
 Ventriculites latiplicatus and radiatus, which is only rarely seen 
 in C. longitudinalis, is always more or less present in this species, 
 and furnishes another characteristic by which it may be at once 
 known from V. mammillaris. 
 
 3. Cephalites paradoxus. PI. XIV. fig. 3. 
 
 Plaits narrow but deep : outer plaits depressed irregularly ; 
 bulging around depressions till the adjoining plaits meet and 
 open into each other : inner plaits regular and simple : pro- 
 'cesses conspicuous : wall thick. 
 
 I have given to this remarkable species the name paradoxus, 
 because it differs from every other species of this genus in having 
 the plaits simple and regular on the inside, while all the com- 
 plexity is on the outside. 
 
 The depressions, and consequently the interspaces between the 
 anastomosing bulgings on the outside, are not of a regular figure, 
 as is the case on the inner surfaces of Ventriculites radiatus and 
 other species. They are varying and elongated; often almost 
 angular ; though, as the plaits are narrow, never very large. There 
 do not appear to be any points of anastomosis* between the ad- 
 joining inner and regular plaits, such as are found between the 
 outer plaits of V. radiatus. 
 
 4. Cephalites alternans. PI. VII. fig. 2, & PI. XIV. figs. 4 & 5. 
 
 Plaits rather broad and very deep : both outer and inner plaits 
 depressed at unequal intervals ; bulging on each side around 
 
 * The appearances seen on dissecting away the inner surface must not 
 be mistaken for this anastomosis. They are, in fact, the bases of the de- 
 pressions on the outer plaits. See the description of a similar appearance 
 on the outside of V . lenuiplicatus, p. 68. 
 
76 
 
 . 
 
 depressions till the adjoining plaits meet and open into each 
 other : processes conspicuous : wall thick. 
 
 The mode of fold in this species resembles that of Ventriculites 
 bicomplicatus in the fact of being repeated on the plaits of each 
 surface. It differs essentially, however, in the fact that the de- 
 pressions, though generally round, are, on neither surface, at re- 
 gular intervals : consequently no regular figure is assumed in the 
 general aspect of either surface. 
 
 I have named the species alternans from the circumstance of 
 the repetition on the two surfaces of the same manner of fold ; 
 while the straight plait is clearly traceable in the central portion 
 of the wall. A transverse section of a specimen of this species is 
 seen on PI. VII. fig. 2. Its difference from a similar section of 
 C. longitudinalis is very marked. 
 
 This is an extremely rare species. 
 
 5. Cephalites bulletins. PI. VII. fig. 3, & PI. XIV. figs. 6 & 7. 
 
 Plaits broad and deep : outer plaits raised in large and very pro- 
 minent projections at considerable intervals, and in such man- 
 ner that they range spirally round the whole body : projections 
 nearly lozenge-shaped and terminating abruptly in an almost 
 flat and somewhat expanded top, having a slight depression from 
 the upper angle towards the middle : inner plaits having large 
 circular depressions at equal intervals ; bulging on each side 
 around depressions till adjoining plaits meet and open into 
 each other : processes very conspicuous : wall very thick. 
 
 This is a most curious and interesting as well as rare but well- 
 marked species. The depressions on the inner folds are much 
 larger than in C. guttatus, which latter have been seen to be 
 larger than in C. longitudinalis. But the external fold is the 
 most deserving of attention. When the specimen is first opened 
 there are seen only a number of nearly semilunar marks. On 
 carefully applying the point of the knife it is found that this se- 
 milunar appearance is caused by very prominent projections, the 
 tops of which are all closed, but have a partial depression at their 
 upper extremity, and which depression is filled as usual with the 
 matrix. The projections themselves are of large size, measuring 
 about two lines in their longest diameter. They stand out nearly 
 or quite half the thickness of the wall, which is generally four 
 lines thick (see fig. 3. PI. VII.*). They differ widely from any- 
 
 * This is a longitudinal section taken rather obliquely in order to preserve 
 the roots. It is not quite regular therefore ; but, on the side which is pre- 
 served, the projections can be well distinguished. 
 
77 
 
 thing we have yet seen. Instead, like Ventriculites mammillaris, 
 of being mere rounded elevations on the plait, they stand out 
 prominently from it ; and a careful dissection shows that their 
 shape is generally that of a lozenge, with the acute angles in the 
 horizontal, and the obtuse in the perpendicular, line of the whole 
 body. Fig. 7* of PI. XIV. shows the manner of the projections 
 from the plait and the figure which the peculiar shape of their 
 tops causes to be seen on a clean section exactly through the 
 middle of any one. In all the specimens of this species which I 
 have seen, the projections run in nearly regular spiral lines round 
 the body. 
 
 On the inner surface of this and of some other broadly de- 
 pressed species there is a very small and slight depression be- 
 tween each of the large depressions, and both on the plaits and 
 on the places of the united bulgings. It is barely traceable, and 
 may easily escape notice. It is however worthy of remark as an 
 additional contrivance for gaining extent of surface, and an ad- 
 ditional instance of the exhaustless variety of plan which nature 
 adopts in the development of life. 
 
 6. Cephalites retrusus. PI. XIV. fig. 8. 
 
 Plaits broad and very deep : outer plaits 
 
 : inner plaits raised in rather small but very promi- 
 nent projections at regular and close intervals, and in such 
 manner that they range spirally round the whole body, and 
 quincuncially relatively to each other : projections cylindrical, 
 rounding off slightly at the top and with an exactly central 
 and rather deep circular depression (sometimes two) on the 
 top of each projection : wall very thick. 
 
 This form departs from every other which has been named. 
 It is the first and only instance in which we find projections on 
 the inner plaits, which have been already more than once found, 
 and will be so again, on the outer plaits. The fold which marked 
 the outer plaits of C. bullatus is here found, with striking modi- 
 fications however, on the inside. The projections are much 
 smaller and closer than in that species, but no less prominent ; 
 while each one is again marked by a deep though small and ex- 
 actly central depression. It is altogether a very extraordinary 
 formf- In chalk specimens it would at once be distinguished 
 
 * In this figure I have connected the inner and outer plaits by brackets, 
 the outline of each merely being given for the sake of clearness. 
 
 f Forms like this afford very strong ground of caution against the hasty 
 adoption of any development theories. The whole of the present subject af- 
 fords, indeed, the strongest ground for such caution. We see infinite variety 
 all subservient to the ends of life ; and throughout which one Unity is 
 
78 
 
 from every other species by presenting, on its inner surface, the 
 appearance of a series of small rings, quite unconnected with each 
 other, but arranged with the utmost regularity. 
 
 It is an extremely rare species. I have only met with a single 
 specimen, and that is a cast of the inner surface in flint, with frag- 
 ments of the characteristic ventriculitic structure preserved in 
 places. I am unable therefore to describe the outer plaits. The 
 characters of the inner ones are, however, so marked that those 
 of the outer ones are quite unnecessary in order to establish the 
 specific difference. 
 
 The name retrusus may be considered either to express the ex- 
 traordinary degree in which the inner plaits are drawn back to 
 form the projections; or that the most marked characters of the 
 species are hidden from external observation by being on the in- 
 ner plaits. In either sense the name seems equally appropriate. 
 
 7. Cephalites catenifer. PL XIV. figs. 9, 14, 15, 16. 
 
 Plaits broad and deep : outer plaits projecting prominently at 
 irregular intervals; projections horse-shoe shaped, with one 
 arm of a lower projection often linked to the hoop of the pro- 
 jection above it on the same plait ; occasional points of ana- 
 stomosis between adjoining plaits : inner plaits having large 
 and generally oval depressions at regular intervals ; bulging 
 on each side around depressions till adjoining plaits meet and 
 open into each other : processes very conspicuous : wall very 
 thick. 
 
 Var. Annulatus. 
 
 Plaits broad and deep : outer plaits projecting prominently at irre- 
 gular intervals; projections ring-shaped, and generally running 
 into each other on the same plait and often anastomosing with 
 those on adjoining plaits so as to form connected rings over the 
 whole surface : inner plaits having large and generally oval 
 depressions at regular intervals ; bulging on each side around 
 
 traceable ; but a Unity which certainly no more points to a low type of orga- 
 nization, or to a necessary or probable progressive development of one form 
 from another, than does the beautiful and philosophical demonstration of the 
 cranial vertebra, or the fact of that demonstration being afforded by the 
 most different members of the Vertebrata. It should be noticed that the 
 very remarkable octahedral structure already developed as characteristic of 
 the membrane of the Ventriculidae has no relation whatever to those "geo- 
 metrical figures " alluded to by Professor Owen. In the present case it 
 is a relative, and not a positive, form ; and one assumed by animal fibre for 
 a special purpose. It has been already remarked (p. 24) that no spicules, 
 or " calcifying salts " enter into the composition of any of the Ventriculidae. 
 See Owen "on the Archetype and Homologies of the Vertebrate Skele- 
 ton," 1848, p. 171. 
 
79 
 
 depressions till adjoining plaits meet and open into each other : 
 processes very conspicuous : wall very thick. 
 
 This is a singular species. The specific name of the typical 
 specimens exactly expresses the appearance of the outer surface, 
 which looks as if several links of a chain were hung about it, 
 sometimes disconnected, often connected, always, or almost 
 always, open on one (and generally the same) side. 
 
 This species will be readily distinguished from C. bullatus by 
 the fact that the semilunar fold is continued down to the upper 
 edge of the plait, as well as by the links being so often con- 
 tinuous, and by each individual projection being much larger. 
 
 Figs. 14, 15 and 16 of PL XIV., all taken from the same spe- 
 cimen, will probably assist in the understanding of this modifica- 
 tion of the fold. Pig. 14 shows a part of the core of the matrix, 
 that which filled the central cavity. The round spots are where de- 
 pressions existed in the body itself, and where, consequently, the 
 matrix projected outwards from the core. Being broken off at 
 each place, these regular marks are left, contrasting strongly with 
 the portions of the membrane adhering to the matrix elsewhere*. 
 Fig. 16, which should be compared with fig. 14 of PI. XIII., 
 shows the peculiar elevations on the plaits : and fig. 15 is a trans- 
 verse section showing three plaits ; the uppermost being struck 
 at a point where there is not any projection, the two others just 
 at the bend of two projections. 
 
 The specimens which I have distinguished as a variety, under 
 the name of annulatus, appear to be cases in which the horse-shoe 
 elevations have become more than usually continuous both on the 
 same plait and by anastomosis with those on adjoining plaits. 
 This character is sometimes seen on the lower part of specimens 
 the upper part of which exhibits the true normal characters of 
 C. catenifer, as in fig. 9. PL XIV. In some cases, however, the 
 same appearance of connected rings, instead of rows of open 
 links, covers a large part, or the whole, of the surface ; and it is 
 important that the true place of such specimens should be un- 
 derstood, whence the utility of distinguishing them as a variety of 
 C. catenifer. 
 
 It generally happens that, even in the most characteristic spe- 
 cimens of this variety, there are places in which the projection 
 on the outer plait stands, as it so often does in the normal C. ca- 
 tenifer, single and wholly unconnected with any other projection 
 on the same or on any adjoining plait. In that case, instead of 
 being horse-shoe shaped, the circle is usually complete. We thus 
 find, on an external plait, a fold very similar to that which cha- 
 racterizes the inner plaits of C. retrusus. 
 
 * See ante, p. 60, note f. 
 
80 
 
 8. Cephalites compressus. PI. XIV. fig. 10. 
 
 Plaits broad and very deep : outer plaits projecting prominently 
 in very elongated loops often linked at one extremity and en- 
 larged at the other : inner plaits often inclining towards, and 
 anastomosing with, adjoining plaits : pouch very short : pro- 
 cesses very conspicuous : wall very thick. 
 
 This appears to be quite a distinct species from the last. Its 
 fold is looser, approaching therein to the character of the group 
 Dilatati. The external modification of fold is very different 
 from that of C. catenifer ; while the internal difference is even 
 more marked. Instead of depressions we have here anastomosis 
 with the adjoining plaits; and at the places of anastomosis 
 the figure becomes almost angular, instead of circular as here- 
 tofore. 
 
 b. Dilatati. 
 
 Head broad and rounding : plaits loose and irregular. 
 
 All the species of the present section differ Very remarkably 
 from the Annulati. In the latter section the heads in all the 
 species were of nearly the same size relatively to the size of the 
 whole body ; as also was the central cavity. The various differ- 
 ences of contrivance by which extent of surface was gained at the 
 same time that the free access of sea-water was maintained, were 
 found in the different modes of folding of the membrane of the 
 wall. In the present section the character of the fold of that 
 membrane differs also in the different species ; but that difference 
 is accompanied by very remarkable differences in the form and 
 extent of the head. The latter becomes the most conspicuous 
 instead of a mere subordinate part to the observer of the whole 
 body. As, therefore, the difference in the heads is a necessary 
 accompaniment of a difference in the fold of the membrane 
 (though rather in the relation of consequence than cause), it will 
 simplify the labour of the inquirer if the character of the head is 
 adopted as one of specific difference. The names given have 
 therefore a reference to this point. 
 
 It will be obvious that, the looser the folds, the more necessary 
 would become the greater extent of cephalic membrane in order 
 to secure the objects already suggested as those for which that 
 remarkable structure was designed. Hence the variations in this 
 conspicuous character in the forms immediately under consider- 
 ation. 
 
 There are minor modifications in individuals of each species 
 which would probably be held by many to justify the assignment 
 of each species as a distinct genus ; an arrangement which would 
 
81 
 
 indeed be far better warranted than many such divisions both in 
 recent and fossil classifications. It does not seeni to me how- 
 ever that the principles of a sound classification will, in the pre- 
 sent state of our knowledge, justify such an arrangement*. 
 Moreover, all the species of the present section are of extreme 
 rarity ; so rare, that it is very probable that few even diligent col- 
 lectors will succeed in obtaining specimens of each, unless some 
 bed abounding in them, and at present unknown, should be dis- 
 covered. 
 
 1. Cephalites capitatus. PL XIV. fig. 11. 
 
 Plaits very deep ; dividing longitudinally, and so reduplicating, 
 very constantly, as they pass from the inner to the outer sur- 
 face; points of anastomosis at irregular distances on both 
 inner and outer surfaces : central cavity small : head rounding 
 and very wide : wall falling in very rapidly but in a regular 
 slope from outer margin of head to root : diameter of- whole 
 body greater than its height. 
 
 In some specimens of the present species the plaits are very 
 traceable on the outside ; in others much less so, on account of 
 the almost total absence of oxide of iron. In each case, however, 
 it is equally obvious that the number of plaits seen on the outer 
 surface is given by the longitudinal division and reduplication of 
 the plaits towards that surface, in the same 
 way as the increase of plaits from base to margin 
 has already been described as being effected 
 by a transverse division and reduplication f. 
 The accompanying figure will explain the pre- 
 sent mode of this reduplication. This arrange- 
 ment takes place to some extent in most of 
 the Annulati, but the very small size of the 
 central cavity in C. capitatus renders this peculiarity constant in 
 this species, and one of its most marked characteristics. 
 
 The general form of this species is so peculiar that a vertical 
 section through the fossil displays a triangular figure, of which 
 the base of the fossil forms an obtuse angle, while the external 
 margins of the head form acute angles with the wall. It is thus 
 impossible to confound this species with C. compressus, as the 
 wall of that species, like that of every other species of the sec- 
 tion Annulati, usually forms, inside and outside, nearly a right 
 angle with the headj. 
 
 * See ante, p. 46 note, and pp. 47, &c. f See ante, p. 64. 
 
 J I have an interesting specimen of this species in which two individuals 
 are close together ; actually touching. But they cannot be mistaken for an 
 example of C. constrictus, each individual having separate roots or places of 
 roots, (see before, p. 51,) and not being parts of one single body. 
 
 F 
 
2. Cephalites cumpanulatus. PL XIV. figs. 12 & 13. 
 
 Plaits very deep ; increasing very rapidly from base, and dividing, 
 and so reduplicating, very constantly, both longitudinally and 
 transversely ; after attaining the fullest expansion, folding 
 inwards and downwards, and gradually contracting till they 
 more or less nearly approach the base, whence, folded upwards 
 in a single plait, the membrane rises, usually simple and 
 plain, in a funnel form, to the margin of the head surrounding 
 the central cavity, to which its marginal edge is attached : head 
 enwrapping the body and attached to all the prominent plaits 
 as far as the point where they incline rapidly towards the base : 
 diameter of body greater than height. 
 
 This species differs very widely in outward appearance from 
 C. capitatus. In that species the head is indeed so largely de- 
 veloped as to be the most conspicuous part of the entire body ; 
 but it still leaves a view to some extent of the wall. In the pre- 
 sent species the whole fossil, unless actually looked at from be- 
 low, is so entirely enveloped by the cephalic membrane, rendered 
 necessary on account of the great depth and consequent tendency 
 to looseness of the fold, that no idea of the character of the 
 membrane of the wall itself can be gained externally. 
 
 The modification of the fold is exceedingly remarkable, and 
 exceedingly difficult to be ascertained. The description given, 
 however extraordinary it may appear, is the result of very laborious 
 and careful examination, comparison, and section of all the spe- 
 cimens which I have been able to obtain. A familiar illustration 
 may perhaps assist in understanding the arrangement of this 
 membrane. If the inquirer will glance at the hangings of any 
 window, looped up, as usual, in festoons at some distance from 
 the ground by curtain pins or ropes, he will see a contrivance 
 rudely imitating the very elegant plan adopted by nature, to give, 
 in a small space, a very great extent of surface combined with 
 security to the polypiferous membrane of C. campanulatus. Take 
 a piece of linen cloth : join together the side-edges along their 
 whole length, gathering the lower edge to a point : fix the upper 
 end of the sac thus formed* to a circular plain wire : at a third of 
 the length from the bottom fix another wire, which, though alto- 
 gether uniting in a circle, is deeply zigzaged : the upper wire 
 remaining fixed, raise the lower wire equally all round, and so 
 that the drapery hanging from the upper simply circular wire 
 
 * To act properly, and to give a full idea of the extent of surface gained, 
 the sac should be very much wider at the middle than at the top or bottom, 
 in order to fill the lower zigzaged wire and yet inclose the plain fold of the 
 cloth without compressing or touching it. 
 
83 
 
 shall fall within, and that hanging from the zigzaged circle 
 shall fall on the outside. Over both wires draw, smoothly, a 
 separate cloth, to which fix both wires. Then, by holding 
 the entire contrivance at any point of the plain wire circle, 
 the whole will be retained in its place. Such a contrivance will 
 afford the best idea of the very remarkable arrangement of the 
 internal membrane of the present species, and of the object and 
 importance of its deeply extended head. It is obvious that, if 
 the lower wire were zigzaged, not only in its horizontal plane but 
 also in a direction perpendicular to that plane, though it would 
 affect the points at which the outer covering or envelope would 
 be touched, it would in no wise affect the principle of the plaits 
 or folds whose extremities touched that envelope. The following 
 two figures may render this matter still clearer. 
 
 In each figure the dotted line is the envelope. Fig. H is a 
 longitudinal section as it would be seen if one could be taken 
 exactly clear through any spot where there was no lateral divi- 
 sion of a plait the presence of which gives a false appearance of 
 anastomosis to a section. It will be seen that the membrane 
 may be traced from the point of the base to c and thence to a 
 in a continuous line, and that the projecting plaits are affixed, at 
 various points, to the envelope a b. Fig. I is a transverse sec- 
 tion taken about the middle of a specimen. The inner circle is 
 the membrane where simple and unwaving, and forming there- 
 fore necessarily an unbroken circle. Between this and the en- 
 velope the plaits are seen, cut horizontally across. 
 
 The membrane begins to fold upwards in its last plait at dif- 
 ferent distances in different specimens from the inner margin of 
 the head ; and it is rarely that the folding upwards will take place 
 on exactly the same plane all around ; whence, on section, a de- 
 ceptive appearance is often given, as if there were a double or 
 triple or still more numerous ramification of the central cavity. 
 
 F2 
 
84 ' 
 
 This is seen in the following curious section of a flint of this 
 species in mv possession. 
 
 & 1 * i re 1 Flff. K. 
 
 bucn cases only afford 
 instances of the care and 
 caution necessary in the 
 investigation of such a 
 subject as the present. 
 
 It will be clearly seen 
 that the mode of fold, of 
 which an attempt has 
 thus been made to convey an idea, secured free access of sea- 
 water to all parts of the surface of the membrane, external and 
 internal. By the same contrivance that membrane was held se- 
 curely in its position ; the regular funnel-shape assumed by the 
 last plait, with its margin fixed at the top, securing it within * ; 
 the campanulate envelope, to which the projecting points of the 
 plaits were affixed, securing it without. The size of the internal 
 cavity varies in different specimens, as will be seen by figs. 12 
 and 13 of PL XIV. ; of which fig. 12 is a general view of the 
 external aspect of one specimen ; fig. 13 is a section, with the 
 matrix cleared away from the inner funnel-shaped simple mem- 
 brane, of another. 
 
 There is certainly no form among the Ventriculidse which 
 might, at first sight, be less supposed than the present to have 
 had any affinity with the fossils which have been described as 
 belonging to the genus Ventriculites. This will be well under- 
 stood by comparing PL XIV. figs. 12 and 13 with any of the 
 figures on PL XIII. 
 
 The condition in which these specimens are found, their deep 
 folds preventing their ever coming free from the chalk, or being 
 developed without the laborious use of the needle, renders it 
 impossible to make any confident observations upon them as to 
 the processes ; a remark which also applies to every other species 
 of the present section. 
 
 3. Cephalites constrictus. PL XV. fig. 1. 
 
 Whole body very low, much-elongated and narrow, with roots at 
 one end : plaits very deep, and running longitudinally from the 
 root extremity; each plait constricted at short and not very 
 regular intervals, and sometimes to nearly its whole depth : 
 cephalic membrane covering the whole upper surface and sides, 
 to the margins of which last, as well as to many of the pro- 
 minent points of the plaits, it is attached; usually constricted 
 
 * See observations on the head, hefore, p. 74. 
 
85 
 
 at considerable intervals, with a single opening in the middle 
 of each compartment thus caused. 
 
 The specific description will satisfy the inquirer that this is a 
 very extraordinary form. Externally it has nothing which would 
 indicate any Ventriculitic affinity, and it has indeed been described 
 by Dr. Mantell under the name of Choanites subrotundus ; but it 
 has no relation whatever to Choanites. The appearance of the 
 fossils is so remarkable, that, but for the fixed rule of preserving 
 every fragment which I could not understand, I should never 
 have been able to establish or even suspect the true affinities. 
 A suite of seventeen specimens enables me, however, now to point 
 out the true general characters of the species without leaving 
 any room even for doubt. 
 
 In the two very different states in which the fossil, or frag- 
 ments of it, are found, it has very different appearances ; the one 
 state (see left hand of figure) shows the upper, the other (see 
 right hand of figure) the lower part only, or its cast. The com- 
 parison of several of these apparently anomalous fossils led me 
 however to conceive that the connected rounded bodies seen in 
 the former set of specimens had some relation to the very pe- 
 culiarly complicated and almost angularly raised surfaces seen in 
 the latter. With this clue I cut down some of these rounded 
 bodies, and found the identical surfaces last named below them. 
 Several sections being made, and the whole series being then 
 compared, order and method became at once apparent where all 
 had previously been anomaly and confusion. The characteristic 
 Ventriculitic structure was detected : the Ventriculitic fold was 
 traced : and the Ventriculitic root Was found. 
 
 I conceive the habit of the animal to have been very different 
 in one respect from that of all the species which have hitherto 
 engaged attention. While the .latter stood rising upwards from 
 a central root, this species, attached at one end by a root, and thus 
 secured in its position, floated horizontally, like a ship riding at 
 anchor. It had therefore no central cavity in the direction of its 
 length, but, instead of this, it was covered by a head investing 
 the upper arid lateral surfaces of that whole length ; and which 
 head, with rare exceptions, for such exceptions do exist, was con- 
 stricted at intervals, causing the animal, when seen from above 
 and entire, as in the greater part of fig. 1. PL XV., to appear 
 like several distinct globose bodies linked together. The fact 
 of the head being occasionally, though rarely, not constricted at 
 all, will satisfy any philosophic inquirer that such an appearance 
 is deceptive, and that the explanation thus given of that appear- 
 ance is the true one. Besides this, however, if the head be re- 
 moved, and the lower surface of the fossil only seen, all trace of 
 separation and distinctness is gone. The membrane of the wall 
 
86 
 
 does not divide into lobes, as in Brachiolites : there is simply, in 
 order to ensure the greater security of the whole polypiferous 
 surface, an occasional constriction of the head and narrowing of 
 the plaits attached to it ; which plaits expand again, like an open 
 fan, in the following compartment. 
 
 The appearance of the plaits themselves is very remarkable. 
 Their frequent constrictions give them a puckered or zigzag ap- 
 pearance, so that a vertical section F ; g L 
 has a figure of this kind. This 
 figure shows, also, how the pro- 
 jecting points of the plaits are 
 often attached, for security, to 
 the head. When the body is broken away the cast left is very 
 curious, the matrix being always broken off in many of the places 
 where it has filled a pucker in the upper plait, depressed where 
 there was a pucker in the lower plait. This is seen on the right 
 hand of fig. 1. PI. XV. 
 
 The species rarely attained half an inch in height or an inch 
 in breadth, though specimens often extend between two and 
 three inches in length. 
 
 It seems to me that the cephalic constrictions most probably 
 mark periods of growth *. They vary in number much in dif- 
 ferent specimens, and, as has been seen, are sometimes not found 
 at all, in which case there are several openings in the undivided 
 head. 
 
 Specimens sometimes assume irregular forms, as if, after 
 death, the long body had become twisted, which I have little 
 doubt was, in many such cases, the real fact. 
 
 I have placed this species next in order to C. campanulatus, 
 inasmuch as, on the one hand, the mode of attachment of the 
 cephalic membrane to the plaits resembles very much that which 
 is found in C. campanulatus, while, on the other hand, the fact of 
 the openings in the head of this species being generally several 
 instead of only one, places it in some relation to the species which 
 will next claim attention. 
 
 4. Cephalites perforatus. PI. XV. fig. 2. 
 
 Plaits wide and very deep, so as to leave no distinct and single 
 central cavity ; dividing, and so reduplicating, very constantly, 
 longitudinally, but not transversely ; somewhat winding both 
 longitudinally and laterally ; occasional points of anastomosis 
 near the outer surface : head covering the entire top and round- 
 
 * Specimens, apparently entire, are sometimes found, Raving one only 
 of the rounded divisions, and thus bearing some resemblance to a very small 
 f. campanula fas, with its root at one end instead of at the base. 
 
87 
 
 ing to some distance down the sides ; having several small 
 round perforations arranged without any regular figure . body 
 of considerable breadth and often nearly twice the height of 
 its diameter. 
 
 The peculiar arrangement of the plaits and head in the last 
 two species rendered any anastomosis of adjoining plaits not 
 essential in either of them. The much greater height of the 
 present species rendered occasional points of anastomosis an im- 
 portant means of securing the permanence of the position of the 
 folds. The width and depth of those folds rendered a large 
 head necessary, while it made unnecessary any large single cen- 
 tral cavity ; the several small openings in the head giving suffi- 
 cient access to the sea-water for the purpose of bathing freely all 
 the internal surface of the polypiferous membrane. The unity 
 of form * is not in the least degree impaired by the existence of 
 these several points of access. The one head still holds in place 
 all the several plaits, a contrivance for the security of the entire 
 animal and of the individual polyps wholly different from that 
 which is found in every species of the family Brachiolites. 
 
 In all the various and so greatly varying forms which have 
 been thus seen to be included in the genus Cephalites one end is 
 found to be subserved, namely, the maintenance of the security 
 of the whole mass, and of each individual of its myriads of living 
 tenants ; together with the unimpeded access of the sea-water 
 that element upon whose constant presence the life and subsist- 
 ence of those myriads depended. The great diversity is no less 
 striking than is, in each case, the completeness of the varying 
 methods which nature has adopted for securing that ever-teeming, 
 ever-active life which excites the inquirer's increased admiration 
 at every step he takes. 
 
 Genus BRACHIOLITES. 
 
 Character. Shape and size very various, but always much lobated 
 or branched : internal cavities of lobes and branches always 
 communicating : extremities closed or open : membrane form- 
 ing the wall sometimes plain sometimes folded : margin of wall 
 thinned or rounded off to an edge: membrane of wall poly- 
 piferous on both external and internal surfaces. 
 
 Departing altogether from the forms hitherto examined, the 
 present genus is characterized by its lobated or branched divi- 
 sionsf. These divisions communicate internally, either by open- 
 
 * See before, p. 58 note. 
 
 f Sec pp. 51, 57 note J, and 85. 
 
88 
 
 ing directly into each other, or by opening into a central cavity, 
 which they surround. 
 
 This genus therefore presents the Ventriculidse in a new light. 
 With far less of the intricate complexity of fold of membrane 
 which is found in the other genera, it exhibits what may, in con- 
 tradistinction, be called a convolution of membrane varying greatly 
 in different species. And that this " convolution " is a distinct 
 thing from the " fold " already noticed will be evident, should its 
 essential distinctness be not otherwise recognized, from the fact 
 that several species have the fold as well as the convolution. In 
 the descriptions which follow, the fold and the convolution will 
 be distinguished as the primary and the brachial fold. 
 
 This genus, like each of the others, will be found to have all 
 its modifications adapted for the purpose of maintaining strength 
 and stability of form and the free access of sea-water. We shall 
 find some contrivances for these purposes of singular novelty and 
 beauty ; and it is upon the marked distinction of two groups in 
 one general arrangement for ensuring to the whole polypiferous 
 surface a full and constant supply of the grand element of the 
 existence of the creatures that the sectional division of the genus 
 is founded; the one section having the separate lobes of such 
 size, or so arranged with reference to a central cavity, that one 
 main* entrance afforded sufficient access to the sea- water; those 
 lobes in the other being so extended, or so arranged with reference 
 to each other, that additional means were needful for that end. 
 
 The roots do not, in general, differ in this genus from the same 
 part in the others ; but I shall have occasion to call attention to 
 some special and very remarkable contrivances in the arrange- 
 ment of this part. 
 
 The forms are all well-defined ; and though, like Ventriculites, 
 specimens of such species as are not of great rarity are found of 
 various sizes, there is little danger of confounding any two of the 
 species unless in a very fragmentary state. As to the question 
 of growth, the present genus may be considered, from the fact 
 just mentioned, to stand on the same ground as the genus Ven- 
 triculites. 
 
 Different forms of this genus are found in the Upper, the 
 Middle, and the Lower Chalk, and even in still lower beds of the 
 cretaceous series. I believe only one, B. digit atus~\ , is strictly 
 
 * The very beautiful arrangement of J5. angularis will be found to be one 
 of those interesting apparent (at first sight) exceptions which prove a rule, 
 the main access being through the central cavity, and the lateral perforations 
 ensuring only the full and free circulation of the water entering, by that main 
 access, the cavities of the arms. The contrivance in all the group Aperti 
 is of a very different character. See post, p. 96. 
 
 t See Ann. and Mag. Nat. Hist. vol. xx. 1st Ser. p. 337, and ante p. 43. 
 
89 
 
 common to all the beds, and that undergoes a modification of 
 character in the lower ones. Some of the most marked forms 
 are, as far as present observation has extended, peculiar to the 
 lower of these beds. 
 
 a. Operti. 
 Brachial folds closed at extremity. 
 
 1. Brachiolites tuberosus. PL XV. fig. 3. 
 
 Membrane having an irregular and generally slight primary fold : 
 brachial fold arranged subspirally around a wide central ca- 
 vity, and at rather distant intervals, in tuberous sacs, broad and 
 flattened at the head, with slight depressions in the middle of 
 the head. 
 
 Of this remarkable species I am fortunate in possessing four 
 well-marked specimens, all of which were found by myself, though 
 in very distant parts of the country, and they are the only spe- 
 cimens I have ever seen. The form is striking. Rising from, 
 apparently, a very short root, it attains a considerable height, one 
 of my specimens being upwards of three inches high. The sacs 
 usually project about four lines from the central cylinder, and 
 are about four lines wide, though sometimes more, at their 
 broadest part. They open by a broad and trumpet-shaped mouth 
 into the central cavity, which is wide and open at the top. Thus 
 the access of sea-water is freely maintained. 
 
 A transverse section gives the accompanying figure, which will 
 be at once distinguished from that seen in a similar section (see 
 fig. G, p. 81) of Cephalites capitatus or of any other of that 
 genus. Fig. M. 
 
 The figure (3. PL XV.), which is 
 from a specimen carefully developed 
 from the chalk by means of the needle, 
 gives a complete idea of the species. A 
 large part is broken away, and thus the 
 inside, as well as the outside, is cleared 
 out and displayed. Being, however, 
 developed by this means, and the spe- 
 cimen being one in which the oxide of 
 iron very greatly abounds, the primary fold is hardly to be seen. 
 A careful comparison of this with other specimens shows that 
 that fold was, like that of Ventriculites impressus, irregular; 
 usually slight ; but occasionally deep, at any rate in that part of 
 the membrane which forms the central cylinder. 
 
 The remark already made, in describing Cephalites campanu-* 
 latus, as to observations upon the processes, applies equally to 
 
90 
 
 every species of the present genus. Enough, however, can be 
 detected to determine the fact of their presence. 
 
 This species must, in its recent state, have been singularly 
 striking. It is difficult, indeed, to conceive anything more beau- 
 tiful than must have been its exterior with each of its regularly 
 ranged lobes covered with its myriads of living polyps and their 
 ever-active tentacles. 
 
 All my specimens are from the Upper Chalk. 
 
 2. Brachiolites elegans. PL XV. fig. 4. 
 
 Membrane simple and without any primary fold : brachial fold 
 beginning almost at the acute base, and rapidly increasing in 
 broad and swelling lobes closely arranged round a central ca- 
 vity, and terminating in a simple and regular crown, open at 
 the top, and which, rising from the midst, reaches to a consi- 
 derable distance above the highest lobes, and is of about half 
 the diameter of the whole body. 
 
 This form is, in its fossil state, frequently so beautiful that I 
 have hence chosen its specific name. The effect is heightened 
 by the fact that the root of this species is usually long, and often 
 maintains the same diameter for a height of nearly two inches. 
 
 The primary membrane is exactly similar to that of Ventri- 
 culites simplex : the brachial folds are deep and broad, their inner 
 surface being freely exposed to access of sea-water from the large 
 internal cavity, The crown is one of the most remarkable fea- 
 tures of this species. It is regular and plain ; springs from the 
 lower edge of the most deeply folded lobes, and rises, unbroken, 
 to a clear and even margin. A glance at fig. H, p. 83, illus- 
 trative of the fold of Cephalites campanulatus, will aid in un- 
 derstanding the anatomy of the present species ; broad convolu- 
 tions here replacing the plaits seen in that specimen, and the 
 crown being always straight-sided, and never, I believe, assuming 
 a funnel form. Nothing can show more convincingly than this 
 peculiar crown, that the forms of the Ventriculidse are riot 
 merely arbitrary massings of an amorphous or simply cumulative 
 organism, or mere examples of " vegetative repetition," but that 
 there was a type appointed to each which it should attain, and 
 each having its special adaptations. In perfect specimens this 
 head is never wanting, though it is rarely, if ever, to be seen 
 without the aid of the knife ; whence it is that it has never, so 
 far as I am aware, been heretofore observed. 
 
 The species appears to be characteristic or the Upper Chalk. 
 
 3. Brachiolites convolutus. PL XV. fig. 5. 
 Membrane coarse in texture, simple and without any primary 
 
91 
 
 fold : brachial fold developed very rapidly from the base, in 
 broad and deeply undulating convolutions, so deep as to leave 
 no regular central cavity; most prominent surfaces much 
 flattened, and spread laterally till those of adjoining convolu- 
 tions meet and unite in many places, often presenting, towards 
 the lower part of the whole, a continuous surface. 
 This species displays many peculiarities. The structure of its 
 simple membrane seems coarse, that is, the squares are larger 
 than in any other species of the Ventriculidse. The root is ex- 
 ceedingly short. The most curious point, however, is the union 
 of the flattened prominences of the convolutions. A reference 
 to fig. M, p. 89, will show, in B. tuber osus, a disposition to 
 flattening of the most prominent part of the convolutions in 
 that species, and will make the nature of the present fold easily 
 understood, where that flattening is so much more extensive 
 that adjoining convolutions meet and unite*. This peculiarity 
 was important in the present species on account of the great 
 depth of the convolutions, which would, without it, have been 
 more liable than in B. eleyans and other species to be displaced, 
 and so injury to have happened to the polyps. But a curious 
 phenomenon is often presented in consequence. The flint, 
 which, from its specific gravity, would always lie, when fluid, 
 near the sea-bottom, was attracted round the base of the spe- 
 cimen ; but, the surface there being usually continuous for a con- 
 siderable distance, access to the lower part of the inside seldom 
 took place, unless the siliceous fluid was very abundant ; any silex 
 attracted towards the inside f in the manner before suggested 
 (p. 13) being specially attracted by the convolutions interposing 
 before it reached the base, solidifying there, and thus preventing 
 the flow of other fluid towards the base. Hence, on the decay 
 of the animal matter, a large hollow was frequently left in the 
 flint, which is often now found only partially filled up by chal- 
 cedony. Such flints are cup- shaped, very regularly rounded be- 
 low, and have a flat top. Near the edge of that top a continuous 
 line is usually found, showing the place of the united mem- 
 brane. Within this are seen traces of the convolutions of the 
 
 * This outer surface, where the polyp-skin is preserved, which very rarely 
 happens, may sometimes appear as if apolypous. A very careful comparison 
 of all the specimens I have, on which remains of this membrane are found, 
 leads me, however, to the conclusion that this is not the fact ; but that, the 
 membrane being coarser than usual, the polyp-cells, though present, are not 
 so clearly preserved as in some other species. 
 
 f The access of the liquid chalk was often made equally difficult by the 
 collapsion, after death, of some of the lower convolutions on one another. 
 Hence we find the bases of specimens enveloped in that substance often hol- 
 low like those in flint. Some deposit of chalcedony has often taken place 
 in those hollows, as in the hollows of the flints. 
 
92 
 
 brachial fold which attracted and retained the siliceous fluid in 
 its descent towards the base. Fig. N is a 
 small specimen of this kind, instances of 
 which are not uncommon in some places. 
 
 Fig. 5. PI. XV. represents a specimen 
 which has been longitudinally divided, 
 and has afterwards had the convolutions 
 cleared out by the needle. The white 
 central parts, round the entire edges of 
 which the structure is seen, are the only 
 parts where the true central cavity is cut, 
 notwithstanding the section. The depth 
 of the convolutions will be clearly seen. 
 It will also be observed that there are several places where the 
 flattened convolutions do not unite, thus leaving ample means 
 for the free admission of sea-water to bathe the whole inner 
 portion of what is really the external surface of the membrane, 
 though so much surrounded by the overspreading flattened, and 
 thus actually outer, surface. The true inner surface is bathed 
 by means of the access of water through the upper part in the 
 usual way. 
 
 Specimens vary greatly in size. I have them from an inch to 
 at least eight inches in height. The height to which the. flat- 
 tened outer surface is continuous varies in different specimens, 
 being, as might be anticipated, greater in large ones. It is some- 
 times much greater on one side than on the other, a circumstance, 
 however, which, in the living creature, would not at all interfere 
 with the free access of the sea-water, inasmuch as the communi- 
 cation was free all around within and under this expanded con- 
 tinuous surface. 
 
 The species is found both in Upper and Middle Chalk. 
 
 4. Brachiolites angular is. 
 
 Membrane exceedingly fine in texture, having a primary fold of 
 minute and corresponding depressions arranged in more or 
 less exact quincuncial figure : brachial fold expanding very ra- 
 pidly into a varying number of arms opening into a central 
 open cavity ; each arm having the two walls parallel and flat ; 
 closed at upper and lateral edges ; terminating at the external 
 angle in a broad triangular lip depressed in the middle ; and 
 having at regular intervals, along its lateral edge, complete 
 perforations, between which distinct and single root-fibres are 
 attached to the membrane at intervals from the base upwards. 
 
 This is certainly the most extraordinary and interesting of the 
 whole family of Ventriculidse. It presents some points of such 
 
93 
 
 remarkable structure as must excite the astonishment and ad- 
 miration of every earnest inquirer. 
 
 Fig. O. 
 
 Few fossils have given rise to more varied conjecture than 
 this. A fragment of a specimen found in an unusual con- 
 dition by Dr. Mantell was originally figured by him under the 
 name of Ventriculites quadrangular is. He subsequently aban- 
 doned that view, showing that it had been adopted without the 
 guide of any definite principle, and, for what reason it is diffi- 
 cult to imagine, and none is stated, placed it among Flustrte*. 
 Others have amused themselves by discovering analogies to the 
 foliaceous sponges. Remains both in chalk and flint, which had 
 apparently escaped Dr. MantelFs notice, or whose connexion 
 with his figured specimens was at any rate not perceived by him, 
 have even been placed by collectors among the Asteria. 
 
 The form in which the species is necessarily most usually found 
 
 * Ante, p. 2, note*. 
 
94 
 
 Fig. P. 
 
 is seen in the accompanying 
 figure. This can, in itself, 
 give little idea of the form or 
 nature of the recent animal. 
 I have a specimen of this 
 aspect nine inches across. 
 Anxious to ascertain the true 
 nature of this remarkable ap- 
 pearance, of which no expla- 
 nation had ever been even 
 offered, I carefully collected 
 every fragment I could find, 
 until I was led to infer a rela- 
 tion between certain flat and 
 unconnected surfaces, some- 
 times found, and these mark- 
 ings. I was fortunate at 
 
 length in obtaining two specimens, each of very large size, and 
 which exhibited on one end the aspect of fig. P, and on the broad 
 side, and in continuous connexion with one of the lines of that 
 peculiar marking, an entire and unbroken surface of many square 
 inches in extent. It became obvious that the inference already 
 made had been correct*; that these markings were caused by 
 the transverse section of a membrane very deeply folded up. I 
 then had recourse, as in other cases, to dissection, in order to 
 ascertain the entire form and habit of the creature, which no 
 specimen developed in the ordinary way could ever show. The 
 extreme fineness of the membrane, and the great depth of the 
 brachial fold, made the task a very difficult one. Having, how- 
 ever, succeeded in several instances, I have been able to restore, 
 from specimens thus cleared out, the beautiful and extraordinary 
 form seen in figure O, a form of animal life seldom if ever exceeded 
 in beauty and striking evidence of design and adaptationf. 
 The wall of this species is exceedingly thin and delicate, the 
 
 * These specimens had been shown by me to several friends, and my in- 
 ference of the habit of the animal, with a model in paper of what I con- 
 ceived its form to be, explained to them long before the importation of the 
 llth livraison of Michelin's ' Iconographie Zoophytologique ' (see ante, 
 p. 8). Plate 30 of that work became an interesting illustration of the pre- 
 sent species, though giving little idea of its true character, and none what- 
 ever of its habit ; on neither of which points do the accompanying descrip- 
 tions afford any real aid. The character of the surface is there very imper- 
 fectly represented, and the magnified views are not truthful, which they 
 could hardly be since that author had no idea of the true structure of this 
 class of fossils. 
 
 f This figure, as well as all the other woodcuts illustrative of the present 
 subject, has been executed, with great care and faithfulness, by Mr. Frederick 
 Gyde. 
 
95 
 
 squares being smaller than in any other species of the Ventri- 
 culidse. Its membrane is folded up in very small depressions, 
 after the manner of Ventriculites quincuncialis, but very much 
 smaller, each depression being rarely more than the sixth of a line 
 in diameter, often less, and usually of an oval form. The plaits 
 may generally be easily traced, as, 'though these folds assume a 
 quincuncial arrangement, it is usually easy, on a large surface, 
 to trace the fan-like expansion of certain lines, the lines of the 
 typical plait. The squares being smaller than in other species 
 of VentriculidaB, the whole thickness of the folded membrane does 
 not exceed the twentieth of an inch. The two surfaces corre- 
 spond as in the section Simplices of the genus Ventriculites. 
 
 Rising from a short and, comparatively to the whole size of 
 the animal, small central root, the fold is sometimes found for a 
 short distance assuming a tubular form, but it generally puts on, 
 very speedily, its characteristic brachial fold into narrow flat tri- 
 angular arms. These folds vary greatly in number. The spe- 
 cimen above figured has ten arms, but that number is very un- 
 common. I have even specimens with only two, but I consider 
 them as abnormal. The arms do not always, though generally, 
 start from exactly the same point. This is the case with the arm 
 on the right of the above figure, as well as with some others the 
 bases of which cannot be seen in the present position of that 
 figure. 
 
 Spreading out, sometimes to a very great, sometimes to a very 
 small size, the angular fold terminates, at the top, not in. double 
 parallel edges as might have been anticipated, but with the edges 
 united and rounded over as at the lateral margin. Each arm 
 is thus closed on all sides except where it communicates with the 
 central cavity, which is, normally, open at the top. The upper 
 and lateral margins of each arm consequently form with each 
 other an acute angle. Those margins are usually straight ; those 
 at the top being also often horizontal, and those at the sides 
 sloping regularly down from their extremity to a point at the 
 base. Instances are found, however, in which these margins 
 have a symmetrical wavy outline. At the extreme outer angle 
 of each arm there is further added a curious triangular expan- 
 sion, not unlike the expansion at the extremity of each sac of 
 B. tuberosus. This lip, as it may be termed, and which, like all 
 the elevations on Ventriculites mammillaris, B. tuberosus, &c., is 
 hollow within, is concave at the aspect towards the spectator. 
 
 It will be understood from this that, between the two walls of 
 each lobe, there is an open space, narrow and of even width 
 throughout. The communication between the central cavity and 
 these arms being therefore not wide, though continuous, the ex- 
 tent and narrowness of these arms struck me for a long time as 
 
96 
 
 likely to offer some difficulty in the way of the free circulation 
 of the sea-water. Confident however that, if my interpretation 
 of .the nature of these animals were correct, the means of free 
 circulation must exist and might be discovered, I resumed the 
 examination with an increased series of specimens. The result 
 was the strengthening of all conclusions as to the physiology 
 of the whole family in general, and of this species in particular, 
 by the discovery of a contrivance by which this end was perfectly 
 effected ; a contrivance which, for its novelty and simplicity, may 
 well claim the inquirer's attention. 
 
 Equidistant, or nearly so, along each lateral margin, I found 
 what at first I took for a larger form of depression, but which, on 
 dissection, I found to be actual perforations* through the mem- 
 brane. These are found present from near the root to the ex- 
 treme angular expansion already named. A comparison of many 
 specimens satisfied me that these perforations are never absent ; 
 that they are always ranged in the same way, and that their size 
 is proportioned to that of the entire animal and to the width be- 
 tween the walls of the arms. They are usually circular, some- 
 times oval. They vary from half a line to two lines in longest 
 diameter, seldom however attaining this last size. 
 
 It became at once evident that this simple provision could 
 have but one end ; but that that end it would fully effect. That 
 end was the fulfilling of the very purpose whose incomplete- 
 ness had before been felt. A constant access and circulation 
 of sea-water would be maintained over all the inner surfaces 
 of these lobes, deep and narrow as they are ; the water, admitted 
 at the opening of the great central cavity, coursing out, by 
 the continual action of numberless ciliated tentacles, through 
 these marginal perforations. And herein it is that these perfo- 
 rations differ from those in the group Aperti, and do not bring 
 this species within that group. The large central opening was 
 amply sufficient, in this case, for the admission of sea-water, but 
 the peculiar form of the arms interposed difficulties in the- way of 
 its free change and circulation, which were equally necessary to 
 the well-being of these creatures. Hence this beautifully simple 
 contrivance for the water, admitted by the central orifice, to pass 
 out through these perforations. In the group Aperti, on the 
 other hand, each perforated lobe offered the principal, if not in 
 every case the only, means of both access and circulation of the 
 
 * In some of Michelin's figures traces of some of these are found, which, 
 when their nature is known, will be recognized, but which, as they appear 
 in those figures, suggest nothing but irregular depressions : they attracted, 
 indeed, no attention of that author, but, in his descriptions, they are wholly 
 unnoticed, the margins being only said to be, in the same words as the 
 " superficiebus laminum," " perforatis ; " which, as applied, is erroneous. 
 
97 
 
 sea-water, just as in any single specimen of Ventriculites OL* 
 Cephalites. 
 
 But the contrivances in this species to secure the well-being 
 of the myriads of its tenants, and so strikingly indicative of de- 
 sign and adaptation, do not end here. Thin and delicate as the 
 wall is, though greatly strengthened by the fact of the upper as 
 well as the lateral margins of each lobe being closed, the cur- 
 rents, small as they would be, caused by the streams continually 
 pouring out of these lateral perforations, or lungs as they may well 
 be called, might tend to displace the form ; a matter of pecu- 
 liar danger in species having such extended surfaces in such 
 close apposition, their distance seldom exceeding one line. 
 Besides this, such broad flat surfaces as the lobes or arms here 
 take would offer so much resistance to the slightest impulse that 
 the walls would be more liable than those of other species to 
 suffer displacement ; while there is no membrane, as in Cephalites, 
 stretching across and attached to each to secure them in posi- 
 tion ; a contrivance indeed which, though so admirably adapted 
 to all cases in which it is found, the length and depth and entire 
 distinctness of the arms would, in this case, have rendered but 
 very imperfectly effectual. 
 
 To secure the animal against such dangers I have further found 
 that, while the central root * is comparatively small, merely acting 
 as an anchor, there depended a single root-fibril from between each 
 of these perforations to a considerable height f^ on all sides of the 
 entire animal. By these then was it maintained securely in shape 
 and in position. These depending fibrils acted exactly like the 
 
 * In an earlier page, 19, contrasting the Actinice, which are locomotive, 
 with the Pennatulida, which are, according to the better opinion, not so, 
 but which are permanently fixed in the soft mud, I suggested that the Ven- 
 triculidae might possibly combine both these qualities, and have a locomotive 
 power, fixing themselves firmly, during pleasure, in the soft mud. However 
 difficult it may be to arrive at certainty on such a point, the inquirer will pro- 
 bably be inclined to acquiesce in the doubtfulness with which this sugges- 
 tion was offered, after considering the very peculiar arrangement of roots in 
 the present species. The roots of this family of compound animals fulfilled 
 the same purpose as the peduncle of the Terebratula. They differ indeed in 
 structure from the latter, for the opportunity of examining which, in the 
 recent state, I am again indebted to the kindness of Prof. Owen, but, in 
 this respect, the byssus of the Pinna, with an analogous function, differs yet 
 again from both. But it is a curious and important fact that the roots of 
 Ventriculidas are never found attached to rock or shells, though shells are 
 often attached to parts of the surface of the body and roots : nor have the 
 minutest shells ever penetrated the substance of the body, as they continu- 
 ally do in sponges. 
 
 f How high it is almost impossible to ascertain. It has been with the 
 greatest difficulty that I have ascertained the fact at all, as it is only by fol- 
 lowing the small fibril with the knife that its presence can be detected or its 
 direction traced. 
 
98 
 
 ropes of a tent, by which alone it is securely kept in a position 
 which, swinging on its single central support, it could not other- 
 wise sustain for an instant*. 
 
 Such are the normal characters of this very curious and inter- 
 esting species. As was to be expected in so delicate a species, 
 specimens are, not unfrequently, found exhibiting abnormal forms; 
 owing either to displacements of the dead mass before or during 
 the process of fossilization ; or, sometimes, to incomplete deve- 
 lopment of the living mass, or accidents to the living animal, 
 accidents to which, from its delicacy and remarkable form, it 
 would be peculiarly liable. Thus I have a flint ten inches in 
 length covered with irregular markings, but in which may I 
 think be clearly traced displaced portions of an individual of this 
 species. I have similar specimens in chalk in which it appears 
 clear that the originally flat arms have been tumbled and bent 
 over upon one another in confusion. I have myself collected 
 more than one specimen which would seem never to have had 
 more than two of the flattened arms. In such cases compensa- 
 ting provisions are found ; and, there being no large central ca- 
 vity, the lateral perforations are found larger than usual, and on 
 both edges of what thus becomes one elongated double fold. I 
 have another most interesting specimen, in which the opening of 
 the central cavity is in an abnormal position, namely on one side ; 
 but still it is there present in all its completeness, thus showing 
 the necessity and the presence of some compensation where any 
 abnormal conditions exist. There is, indeed, no species among 
 the Ventriculidse which bears more conclusive evidence to the 
 truth of the views heretofore expressed as to the character, affi- 
 nities, and habits of the recent animal than does Brachiolites an- 
 gular is. 
 
 I cannot conclude the account of this species without express- 
 ing the strong feeling which its examination impresses of the 
 wonderful variety, and always completeness, of the contrivances 
 by which nature has effected her ever-present purpose of securing 
 the well-being and permanent safety of every creature she has 
 made. The theorists on mere-\ " Vegetative Repetition " and on 
 " Progressive Development " will find themselves equally at fault 
 in the examination of this species. There is no form which the 
 naturalist can study with greater interest, admiration and in- 
 struction than that of Brachiolites angularis. 
 
 This species is found in both Upper and Middle Chalk. 
 
 * The reader will recal the description of an abnormal form of root already 
 given (p. 20). That specimen is of the genus Ventric.ullies. What is there 
 abnormal becomes a special character in the present species. 
 
 f This word is used advisedly, to distinguish those who thus theorize from 
 those who philosophically inquire into the important questions of serial ho- 
 mologies. 
 
99 
 
 b. Aperti. 
 Brachial folds open at extremity. 
 
 1. Brachiolites foliaceus. PL XVI. fig. 1. 
 
 Membrane simple and without any primary fold : brachial fold 
 variously winding and irregularly anastomosing, and thus 
 forming irregular but close and connected sinuous cavities, with 
 rounded, but irregularly arranged, external openings : mass 
 rising to a considerable height ; expanding slowly from root, 
 and the whole maintaining, throughout, a narrow diameter. 
 
 The style of fold of this delicate species is very similar to that 
 of the recent Eschara foliacea. The form assumed by the whole 
 mass differs however materially. Instead of spreading horizon- 
 tally, the habit of B. foliaceus was to rise perpendicularly. It 
 sometimes attained six inches in height, but seldom more than 
 an inch in width at its broadest part. The external openings in 
 this species also differed materially ; those openings being usually 
 separate, and circular or oval in form, not irregularly running 
 into each other as in E. foliacea. 
 
 Fragmentary portions in flint or chalk may readily be distin- 
 guished from those of any preceding species by the greater deli- 
 cacy of the membrane and closeness of the brachial fold. The 
 upper part of fig. 1. PL XVI. shows the appearance displayed on 
 a vertical section ; while the lower part of that figure shows the 
 external appearance of the fossil when entire. 
 
 The close anastomosing of the brachial folds of this species must 
 have given great strength to the whole body ; while the freely 
 communicating cavities would allow constant access and circula- 
 tion of the sea-water. 
 
 I have a specimen in which the whole animal part is converted 
 into iron pyrites and the cavities are perfectly clear of all matrix. 
 
 The species is found in Upper and Middle Chalk. 
 
 2. Brachiolites racemosus. PL XV. fig. 6. 
 
 Membrane having a rather deep primary fold, round, and of 
 nearly equal width the whole depth of the fold, and arranged 
 in quincuncial figure : brachial fold beginning at some distance 
 from the base, and running in narrow and short but regular 
 cylinders ranged subspirally round a small central cavity at 
 rather distant intervals. 
 
 This species differs essentially, from the last in having a deep 
 primary fold ; in having a distinct central cavity, though varying 
 in size, into which each of the brachial folds opens ; and in these 
 
 G 2 
 
100 
 
 last being distinct*, and regular in form and arrangement. The 
 term racemosus seems peculiarly expressive of the character of 
 this species. The perfect animal rose on a rather high stem. 
 I am fortunate in possessing a specimen in flint, from which the 
 fig. 6 on PL XV. is drawn, with the stem and roots entire, a 
 condition in which the members of the present section are very 
 rarely found. I have another specimen in which the processes 
 are very conspicuously seen, also an exceedingly rare circumstance 
 in specimens of this genus. 
 
 The wall of the brachial fold is full a line in thickness, often 
 more, owing to the depth of the primary fold, which much re- 
 sembles that of Ventriculites quincuncialis. The central cavity 
 is usually very small, the access of sea-water being abundant, and 
 its circulation free, by means of the open short cylinders which 
 the brachial fold assumes. Occasionally the central cavity is wide 
 however ; though the opening into it is not, even then, propor- 
 tionably wide, a fact which might have been anticipated. 
 
 It would be difficult to confound this species with B. tuberosus, 
 though the general habit is the same. The primary fold, and 
 the nature of the lobes, open at the extremity, and much larger 
 than those of 5. tuberosus, at once distinguish the twof. 
 
 All the specimens which I liave seen of this species are from 
 the Upper Chalk. 
 
 3. Brachiolites digitatus. PI. XVI. fig. 2. 
 
 Membrane having a deep primary fold of regular quadrilateral 
 and rectangular fornx, usually more or less oblong, and arranged 
 in tessellated figure : brachial fold branching out irregularly 
 
 * I have observed, in one specimen in flint, an indistinct appearance as 
 if two of the cylinders adjoined at one point. If it be a true anastomosis 
 (which I doubt), it is a very rare exception. No symptom of it has been 
 seen in any other instance. 
 
 f Through the kindness of Mr. Wetherell I am in possession of one very 
 interesting specimen of this species, strikingly illustrative of the truth of the 
 views expressed in the early pages of these sheets as to the peculiar state in 
 which silicified specimens are found. Each of the cylinders is separately 
 encased in a thin coat of flint, so that the whole was actually taken, by a 
 distinguished palaeontologist, for the silicified fruit of a conifer. In touch- 
 ing on this subject I cannot forbear citing a passage from Humboldt's 
 4 Cosmos,' published long after my remarks on the formation of flint and on 
 the silicified Ventriculites were written, and which is in direct accordance 
 with, and therefore supports, the view's advanced by me on both those sub- 
 jects. That writer alludes to the siliceous-shelled infusoria as being uni- 
 versally found in sea-water, " although the chemical analysis of sea-water 
 lias not shown silica to be one of its essential constituents ; and it, could only 
 Indeed exist in wafer in a state of simple mixture or suspension." (Vol. i. 
 p. oil.) This, however, is a state of things which Mr, Bowerbank has ex- 
 pressed himself unable, "by any stretch of the imagination," to "eonceivo." 
 4 Ann, and Mag. Nat. Hist'.' vol. xix. 1st Ser. p. 260. 
 
101 
 
 into long and wide regular cylinders either grouped near the 
 base or dividing off one from the other. 
 
 This species differs most essentially from the last. Both pri- 
 mary and brachial fold altogether differ. The former exactly re- 
 sembles the fold of Ventriculites tessellatus. The latter is very 
 peculiar. It often displays a group of cylinders radiating out 
 from near the base just like the outstretched fingers of the hand ; 
 and the length and thickness of the cylinders increase the resem- 
 blance. Hence the name. It is often found however under a 
 modification of this form, rising to a considerable height, and one 
 branch rising out of the other, at considerable distances, as it 
 increases. In each case alike the cylindrical cavities of all the 
 branches open into each other, and there is no true separate 
 central cavity into which they open. The separate branches form 
 exceedingly regular cylinders, and the primary fold is marked 
 on each with perfect regularity. The margin goes off to a round 
 edge, as will be seen in the figure. 
 
 The species cannot be at all understood, or even detected, 
 without careful clearing out with the knife and needle ; since 
 its very nature, like that of every other species in this section, 
 prevents it ever coming out of the matrix entire by any acci- 
 dental fracture : beautiful fragments of it are, however, sometimes 
 found. 
 
 The species is found in the Upper and Middle Chalk. A form 
 essentially the same and of the same habit is sometimes found in 
 the lower chalk and in the chalk marl and greensand*. In all the 
 specimens from these latter beds which I have seen the thickness 
 of the wall is however much greater, and the diameter of the 
 branches also rather greater than in specimens from the Up- 
 per and Middle Chalk, a fact which it is interesting and import- 
 ant to notice as connected with the stratigraphical distribution 
 of these fossils, though I do not conceive these minor characters 
 sufficient to justify, at present at any rate, and until their con- 
 stancy is fully established, a distinct species or even variety for 
 those lower forms. 
 
 Among the Mount Rhanden specimens in the British Museum 
 are individuals identical in general character and habit with the 
 last-named modification of B. digitatus. 
 
 4. Brachiolites tubulatus. PI. XV. fig. 7. 
 Membrane having a more or less slight, but close, primary fold, 
 without any regular figure r brachial fold in narrow tubes in- 
 creasing in size and length from the base upwards, and closely 
 ranged round a central cavity; each tube narrowing at the 
 mouth. 
 
 * See ante, p. 88. 
 
102 
 
 Fig. Q. 
 
 This form will readily be distinguished from every other not only 
 by the difference in the primary fold from that of any species which 
 approach it in the character of the brachial fold, but in the essential 
 character of that brachial fold itself. The narrowness, in com- 
 parison with the length, and the close setting, of the tubes, are 
 together found in no other 
 species. The accompany- 
 ing figure, which is a trans- 
 verse section of a specimen 
 of this species, near the top, 
 and just therefore missing 
 the central cavity, could 
 be presented by a similar 
 section of no other species. 
 It seems to me that there 
 is generally no marked di- 
 stinct opening to the cen- 
 tral cavity, but that it is 
 surrounded on all sides by 
 the tubes. In fig. 7 of 
 PI. XV. the upper part was 
 cut away before it came into my possession, but the wall of the 
 cavity is, in that specimen, rounding inwards, and not expanding. 
 
 It is important to notice that each brachial fold is distinctly 
 tubular and prominently projecting, and that its termination is 
 slightly contracted, in the present species; two characters in which 
 it essentially differs from B. labrosus and.B.fenestratm, and cha- 
 racters which must obviously have materially affected the access 
 and circulation of the sea-water. The central cavity into which 
 those tubes open is another character to which the last remark 
 strongly applies, and is a character also at once distinguishing 
 this species from B. fenestratus. 
 
 This species is from the Middle Chalk. 
 
 5. Brachiolites fenestratus. PL XVI. fig. 3. 
 
 Membrane simple and without any primary fold : brachial fold 
 in narrow tubes anastomosing and opening into each other in 
 not very regular figure, but in several vertical and horizontal 
 planes, leaving interspaces between them about equal to their 
 own width : each tube rounding at mouth and projecting 
 slightly beyond the plane of the most external range of the 
 anastomosed mass. 
 
 The description will at once show wherein this species differs 
 from the last. A single fragment of tube may be mistaken, as 
 the primary fold is not strikingly marked in either, and the size 
 
103 
 
 of the tube is the same in each. But masses of the two cannot 
 be confounded, and the specific distinctions, as noticed in treating 
 of B. tubulatus, are very important. It should be further noticed 
 that, in the present species, there is not that variety in the size 
 of the tubes which, as seen in fig. 7. PI. XV., is found in B. tu- 
 bulatus. 
 
 A certain degree of regularity is generally found, on careful 
 observation, in the arrangement of the anastomosing tubes ; that 
 arrangement bearing often a near resemblance to regular square 
 lattice-work. This arrangement extends both in the horizontal 
 and vertical plane, as endeavoured to be represented in fig. 3, 
 PI. XVI. Specimens sometimes attained a large size. 
 
 Both this species and the last are very beautiful fossils when 
 they can be obtained in any degree of perfectness. This how- 
 ever is extremely difficult, owing to the small branching arms of 
 which each is made up. The inquirer may, unless great care is 
 bestowed, and very cautious dissection made, easily mistake for 
 them some of the markings often found on accidental fracture of 
 chalk and flint, and on the outsides of flints, and which are really 
 caused by B. foliaceus or other of the sinuous species. Oblique 
 fractures take very indeterminate forms. 
 
 This species is found in the Chalk Marl and Upper Greensand. 
 I have never seen a specimen from any higher beds. 
 
 6. Brachiolites labrosus. PI. XVI. fig. 4. 
 
 Membrane having a slight and irregular primary fold : brachial 
 fold variously winding and irregularly anastomosing, thus 
 forming irregular but wide sinuous cavities opening into each 
 other, with slightly projecting wide and irregular openings 
 having entire and rounded margins : mass compact, broad and 
 wide ; rising to a moderate height, and subglobose in form. 
 
 The description will at once enable the inquirer to distinguish 
 this very marked form from every other. The character of the 
 primary fold resembles that of Ventriculites impressus. It is not 
 nearly so close as in B. tubulatus, but much closer than is usual in 
 -B. protensus. . The mouths of the cavities rarely open, as in B. 
 fenestratus, by a regular cylindrical tube, but are very often elon- 
 gated or irregular, and more or less constricted near the middle. 
 The margin, however, is in every case entire, and generally 
 spreads outwards, thick and lip-like, whence the name. The 
 figure exhibits these peculiarities, but the size of the plate did 
 not allow space for the representation of an entire specimen. 
 
 This species is found in the Chalk Marl and in the Upper 
 Greensand. I have seen it from no higher beds. 
 
104 
 
 7. Brachiolites protensus. PI. XVI. fig. 5. 
 
 Membrane having a slight and irregular primary fold : brachial 
 fold in large sinuous tubular masses frequently anastomosing 
 and opening into each other, arid with occasional, but irre- 
 gular, large interstices : mass very irregular and usually spread- 
 ing horizontally. 
 
 This species will perhaps be best understood if the inquirer 
 conceives a number of the arms of B. digitatus to be more or 
 less contorted instead of straight, and to anastomose and open 
 into each other instead of always being distinct from each other 
 at all other points than their bases. The tubular folds of B.pro- 
 tensus project from the mass not very prominently, but still con- 
 spicuously and in every direction. The primary fold differs how- 
 ever, as will be seen both by the description and figure, most 
 essentially from that of B. digitatus. 
 
 The habit of the mass is the very reverse of being compact like 
 B. labrosus ; it may best be described as sprawling, whence the 
 name; its tendency being usually to horizontal rather than 
 perpendicular extension : it does not seem to have had any incli- 
 nation to assume the globose or any other definite general figure. 
 The mouths of the tubes tend to expand as in B. labrosus, w r hile 
 in B. digitatus their tendency, where not simply straight, is rather 
 to contract as in B. tubulatus. 
 
 This species is from the Lower Chalk and Chalk Marl. 
 
 I have thus laid before the reader the result of an investiga- 
 tion which has engaged most of the leisure hours of some years. 
 I am too conscious of the disadvantages under which I labour, 
 and the want of qualifications which I possess, to anticipate 
 otherwise than much criticism as to the results of that investiga- 
 tion and the execution of my task. I would only request the 
 reader to remember that the field was an entirely untrodden one 
 and the task a new one, " a task ono little difficulty in the ac- 
 complishment, and one that may fairly entitle him who enters 
 upon it to expect to meet with indulgence*." 
 
 I have endeavoured to show the existence, in one, at least, of 
 the great geological epochs, of a widely extended class of animals 
 whose nature, if the existence of a few of the forms was vaguely 
 known before, was totally unknown, as also was their structure 
 and all that constitutes the knowledge of an organic being. I 
 have exhibited a structure as remarkable as it is novel. I have 
 shown the extraordinary variety of forms which that structure 
 assumes, a variety in which one Law of Unity, however, still 
 
 * Farre, ut ante, p. ^87. 
 
105 
 
 prevails. Numberless illustrations of design and adaptation 
 have forced themselves upon attention in the course of the inves- 
 tigation, and more might have been suggested had it not been 
 feared that the allusion would appear obtrusive. 
 
 It cannot be supposed that the forms which have been here 
 described constitute all that existed of this family even in the 
 cretaceous seas. In my own collection are a few individuals as 
 to whose specific identity I have some doubt, but as to which I 
 would wait for further means of observation rather than rashly 
 increase the number of species. Doubtless, now that the struc- 
 ture has been described, and figures and descriptions of such 
 numerous forms been given, some attention will be directed to 
 the subject and other forms be found. From the great extent of 
 the materials on which these observations have been made, it 
 may, however, without presumption, be conceived that the prin- 
 cipal typical forms are here included, and that any which may 
 hereafter be clearly ascertained will range themselves easily in 
 one or other of the groups whose characters, general and special, 
 have been here determined. 
 
 Many other obscure fossils are found in the chalk, either gene- 
 rally unknown or distinguished by names which impart little idea 
 of vitality to the objects to which they have been attached, a 
 vitality which it is sought in vain to realize by any descriptions 
 which have hitherto been published. To a more particular ex- 
 amination of some of these, the attention of such readers as have 
 followed with any interest the present inquiry into the structure, 
 affinities and forms of the Ventriculidse, may, at a future day, be 
 perhaps invited. 
 
107 
 
 EXPLANATION OF PLATES, AND INDEX. 
 
 [The Plates are numbered as they were published in the ' Annals and 
 Magazine of Natural History.'] 
 
 PI. VII. (all in flint except figs. 1, 2, 3, 9 and 12). 
 Fig. 1. Transverse section of Cephalites longitudinalis, pp. 17 & 73. 
 
 - 2. Transverse section of C. alternans, pp. 17 & 75. 
 
 3. Oblique section of C. bullatus, p. 76. 
 
 4. Vertical section of Ventriculites, showing base of body lodged and 
 
 ensheathed in the root : the upper part showing the fold of the 
 membrane, p. 16. 
 
 5 & 6. Vertical and transverse sections of the same specimen showing 
 
 body lodged and ensheathed in the roof, p. 19, & see p. 97, note. 
 
 7. Root seen externally ensheathing body, p. 19. 
 
 8. Intimate structure of the Ventriculidae highly magnified, showing 
 
 square and octahedral structure, p. 21. 
 
 - 9. Cast in chalk of this structure, p. 23. 
 
 10. The octahedral structure very highly magnified, p. 22. 
 
 11. The dermis or underskin, pp. 23 & 31. 
 
 - 12. The epidermis or polyp-skin, pp. 23 & 31. This specimen is broken 
 
 away at the upper part, showing traces of the structure below. 
 
 13. The polyp-cells : vertical section highly magnified, pp. 36 & 37. 
 
 14. The mode of addition of fibre, p. 21. 
 
 PI. VIII. (all in chalk except fig. 7). 
 Figs. 4 & 5. Casts showing the places of processes, p. 32. 
 
 6. Polyp-cells highly magnified, p. 36. 
 
 - 7. Root-fibre encrusted with chalcedony, p. 25. 
 
 1. Ventriculites simplex, p. 55. 
 
 2 & 3. Ventriculites impressus, p. 56. 
 
 N.B. For Ventriculites quincuncialis see PI. VII. fig. 7, & p. 58. 
 
 PI. XIII. (all in chalk, except fig. 6, which is in flint). 
 Fig. 1. Ventriculites muricatus, p. 61. 
 
 2. Ventriculites tessellatus, p. 62. 
 
 3 & 4. Different sections of V. tessellatus, p. 62. 
 
 5. Ventriculites cavatus, p. 63. 
 
 6. Ventriculites striatus. p. 63. 
 
 7. Ventriculites mammiltaris, p. 64. 
 
 N.B. For Ventriculites latiplicatus see fig. D, p. 66. 
 
 8. Ventriculites decurrens, p. 66. 
 
 9. Variety tenuiplicatus, p. 66. 
 
 10. Ventriculites radiatus, p. 69. 
 
 11. Section of V. quincuncialis, p. 59. 
 
 12. Section of V. muricatus, p. 61. 
 
 13. Plaits of V. striatus, pp. 63, 65. 
 
 14. Plaits of V. mammillaris, p. 65. 
 
 15. Plaits of V. radiatus, p. 67. 
 
 N.B. For Ventriculites bicomplicatus see fig. E, p. 70. 
 
 PI. XIV. (all in chalk). 
 
 Fig. \. Cephalites longitudinalis, p. 73. 
 2. Cephalites guttatus, p. 74. 
 
 3. Cephalites paradoxus, p. 75. 
 
- 
 
 l"ig$. 4 & 5. Outer and inner surfaces of Cephalites alternans, p. 75. 
 
 - 6. Cephalites bullatus, p. 76. 
 
 - 7. Section' of Cephalites bulletins, p. 76. 
 
 - 8. Cephalites retrusits (moulded from a cast in flint), p. 77. 
 
 9. Cephalites catenifer, p. 78. 
 
 10. Cephalites compressus, p. 80. 
 
 11. Cephalites capitatits, p. 81. 
 
 12. Cephalites campanulatus, p. 82. 
 
 - 13. Vertical section of Cephalites campanulatus, p. 84. 
 
 14. Piece of the matrix from inside of C. catenifer, showing projecting 
 
 parts, which filled the depressions in the living animal, broken 
 away, p. 79. 
 
 15. Transverse section of fold of membrane of C. catenifer, p. 79. 
 
 16. Plaits of C. catenifer, p. 79. 
 
 PI. XV. (all in chalk, except fig. 6, which is flint). 
 Fig. 1. Cephalites constrictus, p. 84. 
 
 2. Cephalites perforates, p. 86. 
 
 3. Brachiolites tuber osus, p. 89. 
 
 4. Brachiolites elegans, p. 90. 
 
 5. Brachiolites convolulus. The specimen has been vertically divided 
 
 and one-half cleared out, thus showing the convolutions and the 
 interior, p. 90. 
 N.B. For Brachiolites angularis see fig. O, p. 92. 
 
 6. Brachiolites racemosus : the right-hand portion shows the form of the 
 
 arms as seen on outside of flint; the left-hand portion shows the 
 root and longitudinal sections of several arms, p. 99. 
 
 7. Brachiolites tubulatus, p. 101. 
 
 PI. XVI. (all in chalk). 
 
 Fig. 1. Brachiolites foliaceus : the lower part showing the outside, the upper 
 
 part a vertical section, p. 99. 
 - 2. Brachiolites digitatus, p. 100. 
 
 3. Brachiolites fenestratus, p. 102. 
 
 4. Brachiolites labrosus, p. 103. 
 
 5. Brachiolites protentus, p. 104. 
 
 ERRATUM. 
 
 At page 17, line 16, a reference is made to PI. VII. by mistake for 
 PI. VIII. 
 
'-. .4W,t- j%. . r,,/. /f,.<t. Voi 
 
 J! 
 
*t. \ ol. 20. 
 
 
 
 r *-# 
 
 J.De C. Jowerbf Jedt. 
 

 
 
 
 / 
 
 / 
 
 
 
 
 
 
Ann .( ^jtfl.y,,f. Hist. S.?. VoLL //, I'// \ 
 
 ' 
 
Ann. S- liaq. Nat Hist. S. 2. VoLl . PL JT 
 
 C.Sowerby.sc. 
 
Aim 3 fay. Nat.JJist. S. 2 . A oil 77. XVI.. 
 
14 DAY USE 
 
 RETURN TO DESK FROM WHICH BORROWED 
 
 EARTH SCIENCES LIBRARY 
 
 This book is due on the last date stamped below, or 
 
 on the date to which renewed. 
 Renewed books are subject to immediate recall. 
 
 ; 
 
 LD 21-50m-4,'63 
 (D6471slO)476 
 
 General Library 
 
 University of California 
 
 Berkeley 
 
I stir 
 
 mm