CHAPTER XXXVI.
The term Cycadophyta, suggested by Nathorst[1069], is used in a comprehensive sense to include both recent Cycads and the much larger number of extinct types which it is customary to speak of as Cycadean plants or fossil Cycads. The designation ‘Cycads’ in the case of the majority of the fossil forms is, however, open to criticism on the ground that they differ too widely from existing genera to be associated with them in one class. It would be pedantic and inconvenient to give up the almost universal practice of extending the term Cycad beyond the limits defined by the characters of recent species. The important point is to adopt some classification which gives expression to our views as to the degree of affinity between recent and extinct types. In a considerable number of cases, especially impressions of presumably Cycadean fronds and stems that occur without any fertile shoots, it is impossible to determine the degree of relationship to modern types. It was with a view to a rational group-designation for such fossils that Nathorst proposed the term Cycadophyta, including in it the two classes Cycadales and Bennettitales. The Cycadales comprise the recent genera and such extinct types as may legitimately be included in the same class, but as we shall see later there are hardly any fossil species that can be assigned to this section on thoroughly satisfactory grounds. The term Bennettitales is used by Engler, Nathorst, and several other authors as a class-designation for a large number of Mesozoic Cycads agreeing in their more important morphological characters with the Lower Cretaceous stems on which Carruthers[1070] founded the genus Bennettites, placing it in a new tribe, the Bennettiteae. Some authors have followed Carruthers in the use of the family-name Bennettiteae for a subdivision of the Cycadales equal in rank to the Cycadaceae, while others, wishing to give greater emphasis to the difference between the extinct and recent plants, prefer to adopt the class-name Bennettitales. Carruthers regarded Bennettites as occupying the same position in relation to other Cycads as Taxus holds with regard to the cone-bearing members of the Coniferae. The adoption of Bennettitales is intended to convey the impression that the class is more distantly related to the recent Cycads than is implied by the analogy of Taxus. Nathorst[1071], as the result of his discovery of certain reproductive organs associated with some Rhaetic fronds described as Dioonites spectabilis, proposed a third subdivision, the Dioonitales, but he subsequently[1072] restored the fronds to their original designation Nilssonia pterophylloides, and on further examination found that the supposed microspore-bearing organs were seeds. Nilssonia pterophylloides would therefore find a more natural place in the class Nilssoniales instituted by Thomas[1073].
BENNETTITALES.
Cycadeoidea. Buckland.
In 1827 Buckland[1074] proposed the generic name Cycadeoidea for some petrified stems from the Purbeck beds in the Isle of Portland and published a description of two species, Cycadeoidea megalophylla and C. microphylla. Brongniart[1075] considered Buckland’s term Cycadeoidea inappropriate and proposed in its place Mantellia, the type-species being Mantellia nidiformis Brongn., the Portland stem which Buckland a month or two later called independently Cycadeoidea megalophylla. Subsequently Brongniart[1076] withdrew Mantellia as it had been previously used by Parkinson for a sponge and substituted Cycadites. Some years later Carruthers[1077] revived Mantellia for a type of Cycadean stem from Portland though one species, apparently indistinguishable from those referred to Mantellia, he named Bennettites portlandicus. There is no adequate reason for the retention of the generic name Mantellia. The close resemblance of the short and thick stems (12–30 cm. in height) described by Buckland to those of some recent Cycads was recognised by Robert Brown and Loddiges and the former suggested to Buckland the inclusion of the fossils in a new family Cycadeoideae. It was this suggestion that led Buckland to adopt Cycadeoidea as a generic name. In a later account of the Portland stems Buckland[1078], in deference to Brongniart’s opinion, substituted Brongniart’s name Cycadites for Cycadeoidea. It is noteworthy that no reference is made in the original description to the occurrence of lateral shoots among the persistent petiole-bases that encase the Portland stems, but in a later account such shoots are represented in one of the figured stems and are compared with the buds occasionally produced on stems of Cycas[1079]. The subsequent researches of Carruthers[1080] demonstrated the reproductive nature of precisely similar lateral shoots in the stem on which he founded the genus Bennettites. The generic name Echinostipes given by Pomel[1081], who had a passion for instituting new nomenclature, to Buckland’s Portland stems has not been adopted: his genus Crossozamia proposed for certain stems and fronds was revived by Carruthers[1082] but has not been generally used. A further complication in the nomenclature of Cycadean stems was introduced by Saporta’s institution[1083] of Bulbopodium and Cylindropodium: in the former genus he included the small ovoid stem figured by Lindley and Hutton as Cycadeoidea pygmaea[1084] and two French Jurassic species which might reasonably be assigned to Cycadeoidea. To Cylindropodium were referred some large French stems from Jurassic strata: an examination of the type-specimens in Paris convinced me that they are typical forms of Cycadeoidea. As Ward[1085] has pointed out, both Saporta’s genera may be merged in Cycadeoidea. Fliche and Zeiller[1086] also include the small globular stems named by Saporta Bulbopodium in the older genus Cycadeoidea: one such type from the Portlandian of Boulogne is described by these authors as Cycadeoidea pumila. Saporta also proposed the name Platylepis[1087], the type-species being Cycadeoidea micromyela Mor., from the Lias of Calvados, which has recently been investigated by Lignier[1088] who wisely adopts Morière’s designation. Another unnecessary generic name is Schizopodium given by Morière to a stem, S. Renaulti[1089], regarded by Lignier as indistinguishable specifically from C. micromyela.
The two names Cycadeoidea and Bennettites have been used by authors for stems which are unquestionably generically identical and as is often the case much confusion has been caused through the failure of palaeobotanists to arrive at an agreement. Lester Ward, a staunch advocate of the rule of priority, repeatedly pointed out that Buckland’s name Cycadeoidea should take precedence of Bennettites on the ground that stems for which these genera were instituted are clearly of the same type, and he added that the older genus, though abandoned by its author, cannot be given up without violating the inexorable rules of priority. On the other hand it has been urged that the genus Bennettites as defined by Carruthers is characterised by the possession of (i) fertile shoots bearing strobili and (ii) vegetative organs exhibiting certain anatomical characters. Buckland’s species of Cycadeoidea, though bearing lateral shoots, are not well enough preserved to afford definite information as to the morphological features of the strobili, nor have we satisfactory data with regard to the degree of resemblance between the vegetative features of Bennettites and the Portland stems. Carruthers laid stress on the elliptical section of the stele in Bennettites in contrast to the cylindrical cylinder of Buckland’s Cycadeoidea trunks. Subsequent research has shown that this difference is not in itself a valid criterion either of generic or specific rank but, as Wieland says, it is an open question to what extent the stem structure of Cycadeoidea and Bennettites agrees. The probability is that were our knowledge of the Portland stems less incomplete, they would be found to possess no anatomical features inconsistent with this generic union. Granting the correctness of this view, the adoption of the later generic term would be a reasonable course to follow on the ground that it stands for stems showing well preserved structure, while Cycadeoidea was in the first instance applied to stems showing only partially preserved stumps of strobili, and the name was not retained by its author. Graf Solms-Laubach[1090] and some other authors have used Bennettites for Cycadean stems possessing strobili of known structure such as those of Bennettites Gibsonianus, reserving Cycadeoidea for similar stems but without strobili sufficiently well preserved to afford evidence of morphological characters. This practice I followed in 1895[1091], but my usage of the two generic names has not been consistent, the name Cycadeoidea being afterwards employed[1092] in a more comprehensive sense. The investigation of the reproductive shoots of American Lower Cretaceous and Upper Jurassic Cycadean stems[1093] has demonstrated the practical identity of their ovulate strobili with those of the English Bennettites. It is clear that the American stems as well as numerous specimens from Italy and other European localities are generically inseparable from Bennettites. The adoption of Cycadeoidea by Wieland, following Ward, has naturally strengthened the claim of Buckland’s genus at least so far as wide usage is concerned, and with some reluctance I propose to fall in with this terminology and for the sake of convenience to give up the use of Bennettites. Among Mesozoic Cycadean stems agreeing with those of recent genera in habit and in the presence of an armour of persistent leaf-bases are several which afford no evidence either of the occurrence of fertile shoots or of lateral shoots with terminal flowers. To such stems the name Cycadeoidea has been applied, the designation Bennettites being restricted to stems with lateral fertile shoots: this practice of treating the absence of reproductive shoots as a characteristic feature of Cycadeoidea is regarded by Wieland as illogical and artificial; and if, as seems probable, he is correct in believing that the majority of the Mesozoic Cycadean plants culminated their vegetative period by prolific development of flowers, the absence of flowering branches would be contingent on a certain stage of development and not a generic character. Dr Stopes[1094] has recently published facts with regard to the anatomical features of Cycadean stems which have a bearing on the vexed question of nomenclature, but are more important from a morphological point of view as they indicate a closer agreement between certain types of Lower Cretaceous stems and those of Cycas and other recent genera, in which successive rings of vascular tissue are developed, than has previously been recognised. In her diagnosis of Cycadeoidea she includes the following statement: ‘In its internal anatomy the trunk shows two or more (up to eight are recorded) zones of secondary wood, the zones composed of distinct series of tracheids each more or less regularly arranged in radial sequence.’ It is claimed that the addition of this character constitutes for the first time a clear distinction between Cycadeoidea and Bennettites. Dr Stopes states that the two stems on which Buckland founded the genus Cycadeoidea are lost and adds that Buckland’s type has certain anatomical features which are not found in Bennettites. Buckland in his description of C. microphylla states that there are ‘two laminated circles’ in the stem instead of the usual single cylinder, a character suggestive of Cycas. In the original figure of this species there are no indications of any lateral fertile shoots though, as Dr Stopes says, many of the Portland stems undoubtedly possessed such flowers. This author makes no reference to Buckland’s later description of C. microphylla: in this it is stated that the stem bears numerous buds rising from the axillae of petioles, and these are shown in the illustration[1095]. Without access to the actual specimen it is impossible to say whether or not the two ‘laminated circles’ described and figured by Buckland[1096] are two distinct cylinders or parts of one cylinder separated by the infiltration of some inorganic substance. In his description of Cycadeoidea Yatesii Carruthers spoke of the occurrence of two cylinders of wood, and this was confirmed in a subsequent account of the type-specimen[1097]. Dr Stopes gives a fuller account of Carruthers’ type and re-confirm a the existence of two vascular cylinders; she also describes a new species, C. buzzardensis[1098] (fig. 578), in which there are 3–8 or more cylinders. Neither of these stems affords any evidence of the possession of fertile shoots; they agree closely with stems of the Bucklandia type in their comparatively slender habit and differ in this respect from Buckland’s Portland species. The occurrence of more than one vascular cylinder in the stems Cycadeoidea (= Bucklandia) Yatesii and C. (= Bucklandia) buzzardensis suggests the possibility that this feature was characteristic of other species included in Bucklandia.
It is clear that some at least of the stems referred to the genus Bucklandia bore flowers of the Williamsonia type[1099], and it is not improbable that the stems described by Dr Stopes as Cycadeoidea Yatesii and C. buzzardensis (fig. 578) possessed fertile shoots comparable with those of the Middle Jurassic species W. gigas. Dr Stopes’s contribution, while establishing a close agreement in anatomical features between some Lower Cretaceous stems and those of Cycas, does not warrant the further conclusion that these stems were in other morphological characters closely allied to modern Cycads. The main features of Cycadeoidea may be summarised as follows: The principal trunk is generally unbranched (fig. 507) and identical in habit with some species of Macrozamia, Dioon, and Encephalartos (cf. figs. 379, 382); in some species, e.g. Cycadeoidea Marshiana, C. superba[1100], C. nana[1101], the plant is represented by several approximately equal, thick, tuberous stems, in some cases easily separated from one another: a similar clustered habit is exhibited by certain forms of Encephalartos. The size of a Cycadeoidea trunk varies from a few centimetres in length with a diameter of similar dimensions, as in C. pumila, to over a metre long as in C. gigantea (fig. 535), or as much as 3–4 metres in C. Jenneyana with a diameter of about half a metre. The surface is covered with persistent leaf-bases, exactly as in many recent Cycads, embedded in a thick mass of ramental scales which often stand out as a prominent reticulum, the petioles having partially decayed before the penetration of the mineralising solution through their harder tissues (fig. 507). On the conical terminal bud preserved in some stems (fig. 536) the foliage-leaves are replaced by linear scales. The leaf-bases often show the vascular bundles which form a simpler system than in the majority of recent species, their arrangement being, as Wieland says, more Fern-like; they form a series of collateral strands following the outline of the petiole-base[1102] with a U-shaped invagination in the middle of the upper face (fig. 517, A). Beyond the structure of the bundles at the base of the fronds and in the cortex of the stem our knowledge of the anatomy of the vascular supply of the rachises is very meagre. At the petiole-base each bundle consists either entirely of radial rows of centrifugal scalariform tracheids and medullary rays usually one-cell broad (fig. 519, B) or of centrifugal and centripetal xylem in varying proportions. The ground-tissue is well supplied with large secretory canals and in the larger leaf-bases there is a considerable development of periderm at the surface (fig. 517, B) as in recent Cycads. The ramenta formed from the epidermal cells consist in most species of fairly broad scales one-cell thick at the edges and broader in the middle; in Cycadeoidea nigra[1103] they are generally one-cell thick throughout and similar ramenta are common in C. Gibsoniana (fig. 517, B). In C. micromyela unicellular hairs replace the scaly ramenta, but transitional forms occur between hairs and scales. In Williamsonia scotica (fig. 562) and in an Indian species of that genus the ramenta are exclusively long hairs as in recent Cycads, the scale-form of the ramenta in Cycadeoidea being a Fern-character. The degree of development of the ramental tissue varies in different species; in C. Stilwelli and C. excelsa it is feebly developed while in Cycadeoidea micromyela the ramenta almost cover the exposed leaf-base armour. The exceptional abundance of the ramental scales is a striking characteristic of some American stems referred to a separate genus, Cycadella[1104]. In the abundance of the ramental tissue, in the compact structure of the well protected cones, and in the thickly cuticularised epidermis of the bracts and leaves Cycadeoidea exhibits xerophilous characters in a very high degree.
With very few exceptions the stems of Cycadeoidea so far described afford no satisfactory evidence of the presence of more than one vascular cylinder in the main stem: in this respect Cycadeoidea agrees with such recent Cycads as Dioon, Ceratozamia, and Stangeria. In the stem of C. Jenneyana the secondary vascular tissue reaches a considerable thickness and shows signs of ‘rings’ in the xylem, but these are probably the result of some periodic interference with the uniform growth and not an indication of more than one xylem-cylinder. The secondary xylem (figs. 508, 509) consists only of centrifugal tracheids with some spiral elements on the inner edge; though manoxylic, Cycadeoidea has more compact wood than that of recent Cycads; the medullary rays are 1–2 cells broad except near the perimedullary region where they are broader and the individual cells are tangentially stretched (fig. 508, B). The tracheids are scalariform, but in C. micromyela, an exceptional type, some of the xylem-elements have 1–2 rows of separate bordered pits (fig. 538). The phloem (fig. 518, B) is characterised by an alternation of thick-walled elements and thinner cells[1105]. From the main stele branches are given off to supply the flowering shoots in the form of a cylindrical stele like that of the primary axis, and each leaf-trace arises as a single bundle which, in the form of a U-shaped strand (fig. 519, A), pursues a straight course to the leaf-base (fig. 510) where it breaks up into several branches (fig. 538, B). Both leaf-trace and peduncle stele are given off from the lower edge of a medullary ray. In the parenchymatous cortex large secretory canals (fig. 518, B) are a conspicuous feature, and accessory cortical steles similar to those in Cycas occasionally occur (fig. 511). The large pith consists of parenchyma and secretory canals, and internal periderm[1106] may be present; no medullary bundles have been discovered.
The correlation of the numerous fronds[1107] preserved in Mesozoic strata with their parent-stems is seldom possible. It is known that leaves of the Zamites type were borne on stems (Bucklandia) agreeing with Cycadeoidea in certain features but differing in the habit of the fertile shoots (fig. 542), and there is evidence that similar stems bore Ptilophyllum and Dictyozamites fronds[1108]; but only unexpanded leaves have been found in actual connexion with Cycadeoidea stems. Wieland[1109] discovered young pinnate fronds, agreeing in the form of the pinnae with Zamites and with some forms of Encephalartos, embedded in a thick mass of ramental scales in the terminal bud of a stem: the rachises were erect and the two-ranked pinnae imbricate and folded inwards (fig. 512), the vernation being like that of Dioon and not circinate as in Cycas and Ferns. The mesophyll of the leathery pinnae is differentiated into palisade and lacunar parenchyma, and the bundles are said to be ‘mesarch’ though on this point more information is desirable. Each bundle is surrounded by a sheath of thick-walled cells and the same tissue forms I-shaped girders as in similar recent leaves. In one species, Cycadella ramentosa[1110], the rachis of a young frond found among the leaf-bases had a broad U-shaped vascular strand.
In most Cycadeoidea stems a characteristic feature is the occurrence of reduced leaves, or bracts, arranged spirally about a cone which sometimes projects slightly beyond the general level of the persistent leaf-bases, or the cones may be fully developed yet still more or less hidden within the armour of petiole-bases (fig. 517, C). These cones, or flowers, are borne at the apex of lateral axillary shoots, and it is characteristic of the genus that they never project more than a very short distance beyond the truncate stumps of the old leaf-bases. The axis of a flowering branch bears numerous linear, hairy, bracts (figs. 513; 514, b) which with their thick felt of ramenta constitute an efficient protective investment. The summit of a fertile shoot forms a rounded hemispherical cushion (fig. 514), or the receptacle may be more elongate and conical (fig. 513) and in some species pyriform[1111]. The flowers are generally bisporangiate, but the fact that in flowers with mature microsporophylls the ovules are small and apparently immature (fig. 513) makes it difficult to determine whether the megasporophylls are merely immature or functionless as in the male flower of Welwitschia (cf. fig. 818). It would seem, however, that in most cases the flowers were bisexual[1112] (fig. 513). A remarkable feature is the enormous number of flowers on a single stem; in Cycadeoidea Dartoni[1113], 54 cm. long and 35 cm. broad, 500–600 flowers were counted on one side. On the other hand the large stem of C. gigantea (fig. 536) shows no flowers. All the flowers on a single trunk are approximately in the same stage of development: it would seem, as Wieland points out, that after a vegetative period of some considerable duration the plant produced a large crop of flowers, and it is conceivable that as in Corypha and some other recent Flowering plants this supreme effort set a limit to the plant’s existence. In the development of flowers from the old stem Cycadeoidea affords a striking example of cauliflory.
The structure of the flowers is one of the most obvious differences between Cycadeoidea and recent Cycads (cf. figs. 393, 394). The bisexual flowers of the fossil stems are comparatively small, not exceeding a few centimetres in length, like a hen’s egg or a medium-sized pear and in marked contrast to the male and female cones of modern Cycads (cf. figs. 393, 394). A verticil of pinnate microsporophylls is attached by a circular collar formed of the concrescent bases of the sporophylls to the lower portion of the receptacle; each pinnule bears several shortly-stalked synangia divided into loculi containing microspores usually with smooth walls and similar, except in their larger size, to those of modern Cycads (fig. 532). Sections of some microspores figured by Wieland[1114] show what appear to be cells in the spore-cavity, but they are larger than ordinary prothallial cells and their true nature is doubtful. We lack information both as regards the nature of the male gametes and the tissue formed on the germination of the spores. The free portions of the microsporophyll-whorl may consist of as many as 20 pinnate sporophylls, but the number is generally smaller; these are much longer than the ovulate part of the flower and in an immature state the ends of the sporophylls are bent inwards and downwards (fig. 513) between the bracts and the receptacle. After expanding and shedding the spores the whole male disc is thrown off leaving a narrow rim below the hemispherical or conical receptacle (fig. 514, d). The sterile distal end of a microsporophyll is generally slightly expanded and spathulate; in C. colossalis[1115] it bears a pair of wing-like appendages (fig. 533).
The upper part of the receptacle is covered with two kinds of organs which together form a layer of uniform depth; these are believed to be homologous, foliar structures; the majority are sterile and have the form of slender long and narrow appendages (fig. 514) each normally with a single axial vascular strand. These so-called interseminal scales increase in breadth in the apical region and form polygonal summits less than a millimetre in breadth or exceeding 2 mm. The low pyramidal apices of the interseminal scales give a mosaic-like appearance to the surface of the ovulate receptacle (fig. 515). In the lower part of the receptacle these sterile sporophylls are the only appendages, but over the greater part of the receptacle they are intermixed with a smaller number of similar appendages which bear a single terminal orthotropous ovule (fig. 514, s). Each ovule-bearing organ or megasporophyll (seed-stalk) is surrounded by 5—6 interseminal scales, and the narrow cylindrical micropylar tubes are readily distinguished from the sterile sporophylls in surface-view (figs. 516, 564). The seeds appear to be exalbuminous and the embryo is dicotyledonous (figs. 516, 521, 523)[1116]. The seeds are very small in comparison with those of living Cycads. The structure of the megasporophylls, both sterile and fertile, is described in more detail in the account of Cycadeoidea Morierei and other forms. Fig. 516 represents a longitudinal section of one of the smallest seeds of Cycadeoidea, between 3 and 4 mm. long; an embryo occupies the greater part of the seed-cavity: the micropylar tube is formed of three layers, a strong inner palisade-layer, a thin middle layer, and an outer palisade-tissue, the interior of the tube being filled with parenchyma; at the shoulder of the seed above the root-end of the embryo the surface of the testa is characterised by 5–6 wings formed by the radial extension of palisade-cells. After the shedding of the spores and the abscission of the microsporophylls the flowering shoot probably increased slightly in length, thus pushing the ripening seeds beyond the ends of the surrounding leaf-bases. The ripe flowers were eventually cut off by an absciss-layer immediately below the receptacle (figs. 521, C, a; 522, a), large cup-like depressions being left on the surface of the stem (fig. 534). The ovulate cone on which the species Cycadeoidea Morierei is founded probably represents the condition of a naturally detached flower of a Cycadeoidea. It is possible that the detached flowers freed from their encircling bracts may have been edible, the small seeds being dispersed by animal agency.
The splendid petrified stem, Cycadeoidea etrusca Cap.[1117], one of the treasures of the Bologna Museum, was found on a tomb 20 miles west of that city where it was doubtless placed by the Etruscans who obtained it from the Upper Jurassic scaly clays in the Apennine Hills. It was in an imperfectly preserved flower of this fossil that Graf Solms-Laubach discovered some microspores, but it was not until Wieland’s examination of the more complete American stems that information was obtained as to the spore-bearing organs. Another Italian stem, Cycadeoidea montiana Cap., was described in 1753 as a ‘congeries of barnacles[1118].’ The specimen named by Goeppert Raumeria Reichenbachiana[1119] was described in the middle of the eighteenth century as a mass of Hippurites or coral cups: this was found in a swamp in Galicia and is now one of the most striking objects in the Dresden Museum; a brief account of it is given on page 409. In 1859 Tyson discovered two large trunks in the Potomac beds of Maryland for which Fontaine subsequently proposed the name Tysonia marylandica[1120], but as Ward pointed out the species is clearly a Cycadeoidea (fig. 507). In 1894 Lester Ward[1121] recorded several additional stems from Tyson’s locality. It is, however, from the Black Hills of Dakota, an isolated spur of the Rocky Mountains, formed of older rocks encircled by Lower Cretaceous strata[1122], that the greatest number of Cycadean trunks have been obtained: the magnificent collection now in the Yale Museum is largely due to the energy and munificence of the late Prof. Marsh. From the Upper Jurassic beds in Carbon Co., Wyoming[1123], several stems have been collected, and a preliminary study of their external features led Ward to institute 20 species of a new genus Cycadella. Stems have also been discovered in the Freezout Hills of Wyoming and additional specimens have been found in the Potomac formation of Maryland. The discovery of over 1000 specimens of Cycadean stems in the Lower Cretaceous and Upper Jurassic beds of a few localities in the United States bears striking testimony to the abundance of these extinct Gymnosperms during the latter part of the Mesozoic era. It is perhaps true that, as Lester Ward wrote, ‘Cycads are to the vegetable kingdom what Dinosaurs are to the animal, each representing the culmination in Mesozoic times of the ruling Dynasties in the life of their age[1124].’ Although the number of stems obtained from European countries is relatively small, the abundance of specimens in the Upper Jurassic strata of the Isle of Portland and Northern Italy and their occurrence in Belgium[1125], France, Germany, Austria, Poland, India and elsewhere, together with an abundance of Cycadean fronds in practically all Jurassic and Lower Cretaceous plant-bearing beds, demonstrate the dominant position of the Bennettitales. The occurrence of the genus Williamsonia in many European localities as well as in India, Afghanistan, and other countries, and the not infrequent occurrence of stems assigned to Bucklandia, Yatesia, Cylindropodium, and other genera furnish evidence of the rich development of the Bennettitales in later Jurassic and Lower Cretaceous floras. Wieland’s recent discovery[1126] of numerous Cycadean fronds, Williamsonia flowers, and stems in Liassic rocks in Mexico shows that the group had already assumed an important position in the early part of the Jurassic period.
The memoir by Carruthers[1127] on British Cycadean stems contains the first account of the morphological features of Bennettitalean flowers based on petrified material. In 1891 Solms-Laubach contributed an important paper on the Bennettitean flower and several years later Lignier[1128] contributed additional data derived from a careful study of a well preserved strobilus[1129] from the Gault of Normandy. Wieland’s researches[1130] added very greatly to our knowledge, particularly with regard to the morphology of the microsporophylls. Among the earlier descriptions of the reproductive organs of the Bennettitales is Buckland’s account[1131] of a specimen from the Inferior Oolite of Charmouth, Dorset (fig. 560), which he referred to the genus Podocarya (= Williamsonia Bucklandi[1132]): a thorough examination of the type-specimen, which unfortunately cannot be found, would probably furnish many important facts.