WWVE.RSI HJLIN01S a L jV\t H URBfW-CHAMPAlCN cr The Ubrary ot tht >.5 3FIELDIANA JAN83 1S/0 1 1 f^ **l^«,», .University ot Illinois I VJl60l09y r' ^a-Chafnn^- Published by Field Museum of Natural History Volume 33, No. 4 September 26, 1974 This volume is dedicated to Dr. Rainer Zangerl The External Morphology of The Inner Ear In Bats From the Phosphorites of Quercy Walter S eg all Research Associate, Vertebrate Anatomy Field Museum of Natural History ABSTRACT Camera lucida drawings of the cochlea and the fenestra cochleae of various species of a number of genera (Pseudorhinolophus, Palaeophyllophora, Paraphyllophora, Vesper- tiliavus, Nycterobius, and Nyctinomus) from the Phosphorites of Quercy have been mea- sured and compared with measurements of the same structures in Recent chiropters. In each case other morphologic details have been noted. The length of the cochlea was measured according to Gray, the width according to my own method. In the majority of chiropters, Recent and fossils, length and width of the cochlea is about equal, although the size of the cochlea differs. These facts let me assume that the shape of the cochlea in chiropters approaches the geometrical figure of an equiangular spiral. In graphs, in which length against width of the fenestra cochleae have been used, representatives of the same species or closely related species are located in close proximity. If two chiropter specimens show a marked difference in the dimensions and shape of the cochlea and the fenestra cochleae, they cannot belong to the same or to closely related species. However, they do belong to the same or closely related species if the dimensions are equal or nearly so. INTRODUCTION In the last century a number of authors (Pictet etal., 1885; Filhol, 1876; Schlosser, 1887-1889) studied fossil bats from the Quercy (Late Eocene- Middle Oligocene). Revilliod in this century, had a great amount of such material from the museums of Basel, Geneva, and Lausanne at his disposal. Among the specimens which he used were a number of complete skulls, a Library of Congress Catalog Card Number: 74-77214 Publication 1191 59 GEOLOGY 60 FIELDIANA: GEOLOGY, VOLUME 33 large number of mandibles, and other bony elements. He published the result of his research in three articles in the memoirs of the Swiss Palaeontological Society (Revilliod, 1917, 1920, 1922). None of the mentioned authors considered the auditory region, except Revilliod and he did it in a very superficial way. The Rhinolophidae are especially numerous in the Phosphorites of Quercy and are represented by the extinct genera Pseudorhinolophus, Palaeophyllophora, and Paraphyllophora and the extant genus Rhinolophus, of which the first two are the most frequent. Three other families appear first in the fossil record in the Quercy deposits. The Emballonuridae with the genus Vespertiliavus; the Vespertilionidae with the genus Nycterobius and the Molossidae with Nyctinomus. ANATOMICAL REMARKS The cochlea in chiropters varies greatly in shape and size. The latter is not related to the weight of the bat (Pye, A. 1967). Since in most microchiropters the cochlea is only loosely connected with the rest of the skull, it is often missing or dislocated in the prepared skulls of Recent and fossil bats. The number of well preserved fossil skulls suitable for my studies was therefore very limited. Preliminary investigations showed that specimens with the same or very similar dimensions and shape of the cochlea and the fenestra cochleae belong to the same or to a closely related species, and this is here investigated in greater detail. The dimensions of the cochlea in microchiropters can be determined relatively easily since the cochlea is covered only by a thin layer of bone. Repeated measurements gave the interesting result that length and width of the cochlea in the majority of specimens are equal or close to equal; only occasionally is there a marked difference between them. D. W. Thompson (1966) refers to Descartes who in 1638 studied the growth of a shell. Thompson says (1966, p. 179): "The shell, like the creature within it, grows in size but does not change its shape; and the existence of this constant relativity of growth or constant similarities of form is of the essence, and may be made the basis of a definition of the equiangular spiral." As the shell grows in size during life, the geometric relationship of the spiral remains constant. The cochlea of the microchiropters is also a spiral and shows a great range of sizes between different species of the same genus. With few exceptions the length/ width ratio is 1.0 or close to it. This suggests that also throughout the microchiropters the cochlea maintains a constant shape at least for adult stages. The left cochlea of Paraphyllophora robusta, Basel Q.P. 1000 (fig. 4) is fortuitously broken and so positioned that the inside of the cochlea can be inspected from above and measurements taken. SEGALL: INNER EAR IN FOSSIL BATS 61 Several radii drawn from the center of the modiolus apparently intersect the whorls of the cochlea at nearly the same angle. This angle is understood to be the angle betweentwo straight lines, the radius vector and the tangent to the curve in the intersecting point. This seems to suggest that the cochlea in the maj ority of chiropters approaches an equiangular spiral. It is of interest that, on one hand, the chiropter cochlea and, on the other, Nautilus should have forms based on spirals with similar geometric properties. The fenestra ovalis (f.o.) of bats is in most instances oval shaped; in some taxa the width of the posterior half is wider than the anterior. Such a shape has also been observed in other orders. Since the variation in the dimensions and the shape of the f.o. in chiropters is rather limited and the f.o. in most fossils poorly defined, it was not used in this study. In contrast to the f.o., the fenestra cochleae (f.c.) shows great variation in size and shape and therefore proves itself an important factor for this study. In a few specimens the inner and the outer rims of the opening of the f.c. differ slightly in their dimensions. In several species of chiropters a bony projection of variable size and shape is present. It originates very close to the posterior edge of the f.c. and extends superficially over part of the latter. This projection was never observed in fossil chiropters but is especially well developed in the Recent genus Taphozous (fig. 5E'). The edges of the f.c. may be very fragile and easily broken in the process of preparation. This can be deceptive and may lead to wrong observations and conclusions. The canaliculus cochleae begins in the scala tympani and leads to the apertura canaliculi cochleae located medial and slightly posterior to the f.c. The apertura externa canaliculi shows variations in shape and size but proved itself difficult to prepare in Recent and still more so in fossil chiropters; it was therefore not used in this study. METHODS In fossil bats rarely is the middle ear preserved. In the species available to me for detailed study the middle ears were lost: inner ears alone were present. These museum specimens had to be kept intact, so that only the external morphology of the inner ear could be studied. It is quite common to find isolated inner ears in collections. This is because in many chiropters they are only loosely connected with the rest of the skull. Generally, they are not used for research because their taxonomy has not been determined. Comparative morphological studies of the inner ear of known taxa may be helpful in the taxonomic identification of such specimens. 62 FIELDIANA: GEOLOGY, VOLUME 33 1mm Fig. l.A, Outline of the ventral aspect of the right cochlea in Pseudorhinolophus, Art 2, Schlosser: zz' circumference of the basal part of the rim of the fenestra cochleae; uu' length of the cochlea; yy' vertical projection of zz'; xx' width of the cochlea. A', Outline of the postero-lateral aspect of the right fenestra cochleae: ww' width of the fenestra cochleae; ll' length of the fenestra cochleae. Camera lucida drawings of a representative series of cochleae were made and from them measurements taken and noted in tables in millimeters. For that purpose the skulls were oriented in such a way that the symmetry plane was vertical and the Frankfurter plane placed in a horizontal position, i.e. normal to the microscopic axis. I used Gray's (1907) method for measuring the length of the cochlea (uu') that is the maximum dimension from the middle (u) of the ventral part of the circumference of the rim (ZZO of the f.c. to the extreme upper end of the cochlea (u') (fig. lA). I measure the width of the cochlea (xx') by first drawing a line (yy') through the vertical projection of the line (ZZ') of the f.c. and then the maximum width (XX') of the cochlea parallel to (yy') was determined. The measurement of the width of the cochlea proved to be more difficult than the measurements of the length and repeated measurements of the width gave slightly different results. Values used are averages of several measurements. Differences in the dimensions of the cochlea and the fenestra cochleae between closely related specimens of the same genus may stem from human error in the classification of the taxon or defects due to the preparation of the specimen. Distortions due to the fossilisation process are considered to be minimal, since obviously distorted specimens were intentionally excluded from the study. SEGALL: INNER EAR IN FOSSIL BATS 63 The materials used were inadequate to enable me to detect or demonstrate any sexual dimorphism. The f.c. (fig. 1 A') is usually wider than long. When using a binocular microscope to view it, and because of the conformation of adjacent bony elements it is often more readily viewed if rotated approximately 90°. However, in the drawings of the f.c. length (LI/) and width (WW) conform to the axis of the head and body. Although the fenestra ovalis (f.o.) is not used in this study, some remarks about it are in place. In an earlier study in which length and width of the f.o. were given as a ratio (=stapedial ratio) marsupials and insectivores were compared. It was shown that most marsupials have a stapedial ratio between 1.1 and 2.1, while in insectivores the ratio varied from 1.8 to 3.0 (Segall, 1970). In both orders, Marsupialia and Insectivora, with specialization of the middle ear, the length of the f.o. increases while the width remains more static. It was found that in Recent bats the ratio of the f.o. varies only between 1.5 and 1.8 with about 1.7 as the mean (Segall, 1971b). The chiropters take up a relatively small area, somewhat between the areas of the marsupials and the insectivores because of the minor variations of the angle of the ossicular functional axis with the Frankfurter horizontal and the small differences in the f.o. ratio (op. cit., fig. 2.), both of which are important factors in the transmission of the soundwaves from the eardrum to the inner ear. One may conclude from it that the middle ear in chiropters is generalized and static. In contrast to the chiropters, gliding mammals, be they marsupials or rodents, have as specialized a middle ear as non-gliding members of the same family. The Dermoptera which have only one genus are all gliders with a specialized middle ear. There are no non-gliding members with which they can be compared. In all three groups the o.f.a. angle with the Frankfurter horizontal is small and the f.o. ratio is rather high (Segall, 1971a,b). These characteristics of the middle ear in gliding mammals provide a strong argument against the theory that gliding may have been a prestage to flying in the origin of bats. As stated above in the great majority of Recent and fossil chiropters the 1/w ratio of the cochlea is 1.0 or close to it, and therefore are not useful in this context. However, the 1/w ratio of the f.c. varies greatly and thus these figures have been used in graphs. HISTORY The greatest number of fossil mammals from the Quercy belong to the family of the Hipposideridae. Schlosser (1887-1889) established the genus MATERIAL FOSSIL1 Institution Catalogue Montauban uncatalogued 1 my reference schlosseri MUnchen 1879 XV 10 C Pseudorphinolophus Basel QP 874 QP852 cf. schlosseri Montpellier UM 5133 Palaeophyllophora oltina Basel QP 784 QP793 Paraphyllophora robusta Basel QP 1000 indeterm 8602 bourguignati Paris 8603 8604, 8605, 8606, 8608 Lyon 8127 Vespertiliavus schlosseri Geneva 81018, 81020 wingei Montauban uncatalogued 4 my reference gracilis Montpellier UM 5135 Nycterobius gracilis Basel QP632 Nyctinomus stehlini RECENT Basel SG 6240 All Recent specimens are taken from the mammal collection of the Field Museum of Natural History (FMNH) and are listed in the tables. 1 Fossils are indicated in the graphs with *. Table I. Dimensions of the cochlea and the fenestra cochleae and the ratios (length/ width) of the fenestra ovalis in four specimens of Pseudorhinolophus. Catalogue Cochlea Fen. cochleae Fen. o\ Institution no. Side length width length width 1/w Ps. Ait 2 of Schlosser MUnchen 1879 XV 10C r 3.25 3.25 0.4 0.9 1.7 Ps. schlosseri Basel Q.P. 852 r 1 2.5 2.58 2.3 2.25 0.4 0.4 0.64 0.625 1.7 Ps. schlosseri Basel Q.P. 274 1 2.5 2.5 0.4 0.6 Ps. cf. Montauban uncatalogued 1 2.91 2.916 0.32 0.6 1.7 schlosseri 1 my reference 64 SEGALL: INNER EAR IN FOSSIL BATS 65 Pseudorhinolophus and included within this genus Rhinolophus antiquus Filhol (1872) and Vespertilio morloti Pictet (1855). Revilliod (1917), who had much Quercy material at his disposal, was able to distinguish on the basis of the length of the dental structures two genera, Pseudorhinolophus, the most numerous among the Quercy mammals, and Palaeophyllophora, both belonging to the family Hipposideridae. Pseudorhinolophus Schlosser (1888) Revilliod established four species in the genus Pseudorhinolophus: Ps. (Vespertilio) morloti (Pictet); Ps. schlosseri; Ps. weithoferi; and Ps. egerkingensis. The genus exists in the Lutetien superior of Egerkingen and in the Bartonien-Ludien of Mormont (Dechasaux, 1958). Revilliod concluded that Pseudorhinolophus is a genus of the family Hipposideridae which became detached from the main trunk earlier than the genus Hipposideros. Some recent Hipposideros are similar to Pseudorhinolophus, others to Brachipposideros of a later age (Bouzigues). Sige (1968) has recently included Pseudorhinolophus and Brachipposideros subgenera within the genus Hipposideros and he reports that H. (Ps.) bouziguensis, the most frequent mammal on the border of Oligocene and Miocene, is the last representative of the subgenus Pseudorhinolophus. The Ps. skull (Art 2) pictured by Schlosser ( 1887- 1 889) has a cochlea of larger dimensions than that of one of the skulls of Ps. schlosseri (Q.U . 852) Fig. 2. Outline of the basal view of the cochlea and the postero-lateral aspect of the fenestra cochleae in A and A', Pseudorhinolophus, Art 2 of Schlosser, Munchen, 1879 XV 10c; B and B\ Pseudorhinolophus schlosseri, Revilliod 1917, Basel Q.P. 852; C and C, Pseudorhinolophus cf. schlosseri, no. I my reference Montauban; D and D', Hipposideros commersoni FMNH 95153; E and E\ Hipposideros diadema griseus FMNH 80367. 1.0 .5 2 .1 length * 1879 XV 10c Ps. sp. nov. (Art #2 of Schlosser) Ps. cf. schlosseri "A" #1 Ps. schlosseri .2 .3 J .6 mm Graph I. Length against width of the fenestra cochleae in several Pseudorhinolophus specimens. 66 SEGALL: INNER EAR IN FOSSIL BATS 67 Table 2. Available dimensions of the cochlea and the fenestra cochleae in some Recent Hipposideros. cochlea fenestra cochleae catalogue no. (FMNH) length width length width Hipposideros commersoni 95153 4.0 4.0 0.44 1.0 H. commersoni 95152 4.0 4.0 0.44 1.0 H. commersoni merungensis 67928 3.75 3.75 0.56 0.92 H. diadema griseus 80367 3.75 3.7 0.24 0.68 H. diadema vicareus 76960 3.91 3.5 0.28 0.68 H. diadema 1118 3.75 3.75 0.28 0.68 H. armiger 39537 3.66 3.66 0.24 0.64 H. armiger 39540 3.58 3.58 0.2 0.6 H. coffer 104522 0.28 0.52 H. coffer 104532 0.28 0.52 H. coffer 104521 2.41 2.41 0.32 0.52 H. larvatus 82660 2.66 2.58 0.2 0.48 H. larvatus 32109 2.66 2.58 0.2 0.48 H. cineraceus 82706 0.16 0.28 H. cineraceus 32224 0.16 0.28 H. cervinus 31617 2.6 2.15 0.2 0.36 H. cervinus 31618 2.17 2.17 0.2 0.4 used by Revilliod (compare figs. 2A, B and table 1). The width of thef.c. in the Schlosser specimen (fig. 2A') is greater than in Revilliod's specimen (fig. 2W) but the length of the f.c. is equal in both. The Ps. cf. schlosseri from the Montauban collection (1 of my reference) has a cochlea of a size between Schlosser's and Revilliod's specimens. Although there is a marked difference in the dimensions of the cochlea (c.) and f.c. between these three specimens there is no difference in the ratio of the f.o.; it is in all Pseudorhinolophus specimens studied 1.7. The small variations of the f.o. in chiropters has been pointed out already in a former paper (Segall, 1971b). Revilliod found some similarities between Pseudorhinolophus, on one hand, and Hipposideros commersoni and H. diadema, on the other. The similarities resp. dissimilarities in the dimensions of the cochlea and the fenestra cochleae between these taxa are shown in Tables 1 and 2, Figure 2 and in the Graphs 1 and 2. Palaeophyllophora Revilliod (1917) This is the second genus which Revilliod (1917) established among the fossil Hipposideridae from the Quercy. It is known at least since the Bartonien-Ludien because one finds it in the Mormont which is of that age (Dechasaux, 1958). It does not exist after the Quercy and has no Recent l.Or- 9- .8 - .7- .6- .5 .1 H. commersoni •39540 39537 J H. armiger H. caffer 104522 m 10452 104532* •1045^1 82660. 32109* H. larvatus • 31618 • 31617 H. cervinus -82706 732224 \ H.cineraceus 67928 length .3 .5 .6 mi Graph 2. Length against width of the fenestra cochleae in several species of Hipposideros. 68 SEGALL: INNER EAR IN FOSSIL BATS 69 Fig. 3. Outline of A the basal view of the cochlea and A' the posterolateral view of the left fenestra cochleae in Palaeophyllophora oltina, Basel Q.P. 793. representatives. Revilliod (1917) distinguished two species: P. quercyi and P. St. Nebulae. The latter was recognized (Revilliod, 1922) to be a synonym of P. oltina (Delfortrie). Two P. oltina, both from the Basel collection were available for my study (fig. 3). The cochlea of Pseudorhinolophus used by Schlosser (table 1) is about the same size as that of Palaeophyllophora oltina, while Pseudo- rhinolophus used by Revilliod has a smaller cochlea. The width of the f.c. is smaller in Ps. schlosseri than in P. oltina while the length is about equal in both. Paraph vllophora Revilliod (1922) Revilliod (1922) established the genus Paraphyllophora and the species P. robusta. The left cochlea of this specimen from the Basel collection (Q.P. 1000) is clearly not in situ; the right appears to be in situ but is broken; therefore exact measurements can not be taken. Width and length are about 3.3 mm. (fig. 4). Vespertiliavus Schlosser (1888) Filhol (1876) had an incomplete upper jaw of a specimen which he called Vespertilio bourguignati. It differed from the then known genera Table 3. Dimensions of the cochlea and the fenestra cochleae in two specimens of Palaeophyllophora oltina. cocf ilea fenestra cochleae Catalogue no. Institution side length width length width 784 Basel r 3.5 3.5 0.44 1.16 1 3.5 3.5 0.44 1.16 793 Basel r 3.41 3.41 0.52 1.16 1 3.41 3.41 0.48 1.1 1.2 r 1.1 1.0 length 3 .6 mm Graph 3. Length against width of the fenestra cochleae in two specimens of Palaeophyllophora oltina. Fig 4. Photo of a basal view of both cochleae in Paraphyllophora robusta, Basel Q.P. 1000; Right cochlea in situ, approximately its basal half is broken off; left cochlea dislocated, its tip broken off. 70 SEGALL: INNER EAR IN FOSSIL BATS 71 from the Phosphorites at his disposal. Schlosser (1888) who studied six lower jaws which also differed from the then known genera of the Phosphorites, assigned them to a new genus Vespertiliavus. He also included the Filhol specimen of Vespertilio bourguignati within his new genus. However, beyond Ve. bourguignati he did not name any of these specifically. Instead he indicated that to judge from their size differences they belong to four species, and he was not certain which, if any of them, were to be correlated with Ve. bourguignati, the genotype species. The new genus Vespertiliavus was primarily based on the dentition differences from Vespertilio. According to Revilliod (1920), Vespertiliavus is a palaeontological precursor of the Emballonuridae which derive from a very primitive group; the Messel forms are representatives of this ancestral group (Russell and Sige, 1970). Revilliod accepted Ve. bourguignati (Filhol) and added to it the materials of Schlosser species 2. He also named three new species of the genus Vespertiliavus: Ve. wingei, Ve. schlosseri, and Ve. gracilis. All taxonomic arrangements were based on measurements of maxillae and mandibulae. Four Ve. bourguignati, three Ve. schlosseri, one Ve. gracilis, one Ve. wingei, and one undetermined species were at my disposal (see fig. 5 and table 4). Table 4. Dimensions of the cochlea and the fenestra cochleae in three species of Vespertiliavus. cochlea fenestra cochleae species institution catalogue no. side length width length width Ve. sp. indeter- mined Paris 8602 dext. 3.58 3.34 0.84 1.28 Ve. bourguignati Paris 8603 sin. 3.4 3.75 0.64 1.04 Ve. bourguignati Paris 8604 sin. 3.4 3.4 0.6 1.08 Ve. bourguignati Paris 8605 sin. 3.4 3.4 Ve. bourguignati Paris 8606 sin. 3.41 3.41 Ve. bourguignati Paris 8608 sin 3.4 3.4 0.6 1.12 Ve. schlosseri Geneva 81019 sin. 3.0 3.0 0.56 1.04 Ve. schlosseri Geneva 81020 sin. 3.0 3.0 0.52 1.0 Ve. schlosseri Lyon 8127 sin. 3.0 3.0 0.56 0.92 Ve. wingei Montauban uncatalogued 4 my reference sin. 3.416 Ve. gracilis Montpellier 5135 dext. 2.91 2.85 Table 4 shows similarity in the dimensions of the cochlea, slightly less in those of the f.c, between specimens of Ve. bourguignati; part of the 72 FIELDIANA: GEOLOGY, VOLUME 33 differences in the f.c. can be explained by small uncertainties in the measures. The dimensions of the cochlea in Ve. wingei hardly differ from those in Ve. bourguignati. In both, the cochlea is larger than in Ve. schlosseri and Ve. gracilis. However, the f.c. in Ve. bourguignati has similar dimensions as in Ve. schlosseri. The dimensions of the cochlea and f.c. in the Ve. indet. (8602) are larger than in all the above mentioned species; it could be the representative of a different species of Vespertiliavus. The similarity of Vespertiliavus with Taphozous, noted by Schlosser and later by Revilliod, was confirmed by my comparative studies, especially a similarity with Taphozous liponycteris (tables 4, 5; graphs 4, 5; fig. 5). Revilliod states further that in the characters of skull and dentition Vespertiliavus is much more primitive than the Recent Emballonurids and that it is a phylum which has developed in parallel with the other Emballonurids similar to the relationship of Palaeophyllophora with Recent Hipposideridae. However, one cannot place Vespertiliavus in the Table 5. Dimensions of the cochlea and the fenestra cochleae in some Recent Em- ballonuridae. FMNH cochlea fenestra cochleae catalogue no. length width length width Emballonura semicaudata 31659 2.41 2.0 0.36 0.6 E. semicaudata 31651 2.4 2.0 0.36 0.6 E. monticola 77010 2.25 2.25 0.28 0.52 E. meeki 54080 2.25 2.25 0.4 0.68 E. meeki 54078 2.25 2.25 0.4 0.68 E. nigrescens 8249 2.25 2.0 0.2 0.63 Taphozous melanopogon 32257 3.08 3.0 0.56 0.96 T. melanopogon 98694 2.91 3.0 0.64 1.0 T. perforatus 67550 2.75 2.75 0.64 1.04 T. liponycteris 82785 3.25 3.17 0.64 1.08 T. liponycteris 82784 3.0 3.0 0.6 1.12 Coleura gallarum 67344 2.58 2.5 0.4 0.76 C. gallarum 67345 2.58 2.58 0.4 0.76 C. gallarum 67348 2.5 2.5 0.46 0.76 C. afra 78178 2.57 2.57 0.48 0.88 C. afra 78179 2.58 2.58 0.48 0.88 Rhynchonycteris naso 43118 2.08 2.0 0.14 0.48 R. naso 21993 2.16 1.91 0.14 0.48 R. naso 62136 0.16 0.52 Saccopteryx bilineata 19213 2.5 2.5 0.32 0.84 S. bilineata 68340 2.58 2.5 0.32 0.84 S. bilineata 19211 0.28 0.84 S. canescens 18707 2.16 2.16 0.28 0.7 S. leptura 43122 0.32 0.76 S. leptura 21659 0.36 0.8 SEGALL: INNER EAR IN FOSSIL BATS 73 13 r- 12 11 10 Ve. sp. indetermined length J 9mm Graph 4. Length against width of the fenestra cochleae in three species of Vespertiliavus. direct ascendancy with the Recent Taphozous or any other genus of the Emballonurinae by comparing the skull only. In Taphozous there is a process of the periotic which projects over a part of the f.c. (fig. 5E'); this process appears not to be present in Ves- pertiliavus. Nycterobius Revilliod (1922) Revilliod (1922) established a new genus Nycterobius and the species N. gracilis which he included in the family Vespertilionidae. The specimen which he studied came from the Phosphorites of St. Neboule Savin, near Cajarc. I had the opportunity to study this specimen (Q. P. 632) of the Basel Collection. Revilliod noticed several similarities between the skull of N. gracilis and some species of the small Myotis group, especially with Myotis capaccinii, on one hand, and also with Chilonatalis tumidifrons, a natalid, on the other. He came to the conclusion that N. gracilis is closer to M. capacinii than to Ch. tumidifrons. My comparative studies of the inner ear agree with Revilliod's findings (table 6, fig. 6, graph 6). The 74 FIELDIANA: GEOLOGY, VOLUME 33 Fig. 5. Outline of the basal view of the cochlea and the postero-lateral view of the fenestra cochleae in A and A', Vespertiliavus spec, indet. Paris 8602; B and B\ Vespertiliavus bourguignati Paris 8604; C and C, Vespertiliavus schlosseri Geneva 81020; D and D\ Emballonura semkaudata FMNH 31651; E and E', Taphozous liponycteris nudiventris FMNH 79567. dimensions of the cochlea in N. gracilis and M. capaccinii are quite similar. This applies also to the f.c. when the inner opening of the f.c. in M. capaccinii (FMNH 44105) is considered. The outer opening is slightly larger (table 6). Revilliod also pointed out differences in the skull between the three taxa. Since my studies show that specimens with the same or very similar dimensions and shape of the cochlea and the fenestra cochlea belong to the same or to a closely related species, I conclude that N. gracilis and M. capaccinii are closely related. Both specimens come from southern Europe, N. gracilis from the Quercy and M. capaccinii from Lugano, Italy. It would be interesting to see how other parts of the skeleton compare. Nyctinomus Revilliod (1920) Revilliod described the genus Nyctinomus and the species Ny. stehlini which was recovered from Montaigu-le-Blin, north of St. Gerand-le-Puy. He found in the form and size of the skull similarities with small Recent Molossidae, especially with Mormopterus. The cochlea and the fenestra cochleae were drawn and measured in a number of Molossidae. The results are shown in the Table 7, Graph 7 and Figure 7. DISCUSSION AND CONCLUSION The cochlear lengths and widths have been measured. In the majority of the chiropters studied length and width of the cochlea are equal or close to u,- 1.1 1.0 .6- 3 ■a Taphozous (#82784 \ T. liponycteris V \ — J -V^t2785 , T. perforatus 087550 • 98694 / T. melanopogon Saccopteryx • 19211* I C. af ra - — X \ •' V 78178 H \ / • 78179 >lffi3' / Coleura ' 87345 S. bilineata / /-.-..'i^vi' /SIxi* 'r> •■ — , — S 218599 '/ /87344 /C. gallarum S. leptura, f I \ ' S. canescens , K E. meeki 18707* s^'/ZUoio, E. nigrescens ^8249 /'~31659 / EmbaMonura \ (#3: "31551 E. semicaudata • 82135\ -„A,n \ #77010 2i993 43118 Rhynchonycteris naso ■' E. monticola length .1 J .7mm Graph 5. Length against width of the fenestra cochleae in several Recent Emballonurinae. 75 lOr Myotis / 7 M.mystacinus 98713 and M.gracilis 47798 \ M. ricketi 43311 -t M. capaccinii 46016,44015 / M.chiloensis 63827,73737 Nyctiellus lepidus 44536, 44544 \ M. albescens 68471,68474,68476,41512 Chilonatalus tumidifrons 34069 Natalus I V^>y M. siligorensis 33857 , "\ IM. mexicanus 20103,49996 / / N. stramineus 65638 /■-7 length J .6 mm Graph 6. Length against width of the fenestra cochleae in Nycterobius gracilis compared with several Recent species representing Myotis (Vespertilionidae), Natalus, and Nyctiellus (Natalidae). Fig. 6. Outline of the basal view of the cochlea and postero-lateral view of the fenestra cochleae in A and A', Nycterobius gracilis Basel Q.P. 632; B and B', Myotis capaccinii FMNH 44105 and C and C, Chilonatalus tumidifrons FMNH 34069. 76 SEGALL: INNER EAR IN FOSSIL BATS 77 Table 6. Available dimensions of Nycterobius gracilis and some Recent Vespertilionidae and Natalidae. Nycterobius gracilis r 1 Myotis capaccinii M. capaccinii M. ricketi M. chiloensis M. chiloensis r 1 M. albescens M. albescens M. albescens M. albescens M. mystacinus M. gracilis M. siligorensis Natalus stramineus N. stramineus N. mexicanus N. mexicanus Chilonatalus tumidifrons Nyctiellus lepidus N. lepidus N. lepidus * Inner rim of fenestra cochleae ** Outer rim of the fenestra cochleae it and therefore the ratio is constant. This lets me assume that the chiropter cochlea approaches the geometric shape of an equiangular spiral. The fenestra cochleae in the different genera, however, behaves differently. The length/ width ratios are much more varied; while the width measures vary considerably, corresponding length measurements do not. Therefore the f.c. ratio was more useful than the cochlear ratio in this study. If two chiropter specimens have cochlear and fenestra cochleae dimensions and shapes which are identical or close to it, it is very probable that they belong to the same species or to closely related species of the same genus. I have obtained this result in a wide sampling of the dimensions of these structures in a number of families, genera, and species. This result is not surprising, when one considers the importance of the cochlea with the fenestra cochleae in chiropters. For instance, several specimens of Vesper tiliavus bourguignati were studied (table 4) and the dimensions of their cochleae and fenestrae cochleae were found to be quite similar. FMNH cochlea fenestra cochleae catalogue no. length width length width Q.P. 632 2.16 2.16 0.2 0.16 0.64 0.64 44105 2.16 2.16 0.24 0.24 0.64* 0.8 ** 44106 2.16 2.16 0.24 0.72 43311 2.25 2.25 0.28 0.92 73737 1.91 1.91 0.28 0.72 63827 1.91 1.91 0.28 0.72 41512 1.91 1.91 0.2 0.64 68476 1.91 1.91 0.2 0.64 68474 0.24 0.68 68471 0.28 0.68 98713 1.83 1.83 0.24 0.68 47798 1.83 1.83 0.24 0.68 33857 1.66 1.66 0.2 0.6 51160 1.91 65638 1.91 1.91 0.16 0.48 20103 0.2 0.56 49996 0.2 0.56 34069 1.84 1.84 0.24 0.64 44539 1.75 1.75 44544 0.32 0.68 44536 0.28 0.68 Table 7. Available dimensions of the cochlea and the fenestra cochleae in Nyctinomus stehlini and some Recent Molossids. catalogue cochlea fenestra cochleae institution no. length width length width Nyctinomus stehlini Basel S.G.6240 2.41 2.41 0.56 0.92 Mormopterus kalinowskii FMNH 19952 2.5 2.5 0.56 0.88 Tadarida cynocephala FMNH 7867 2.58 2.5 0.56 0.72 T. cynocephala r 1 FMNH 7868 0.56 0.56 0.76 0.76 T. cynocephala FMNH 7869 0.56 0.76 T. mexicana r 1 FMNH 47858 0.6 0.6 0.88 0.84 T. mexicana r FMNH 1069 2.41 2.41 0.6 0.92 1 2.46 2.41 0.6 0.92 T. mexicana FMNH 13273 2.41 2.41 T. mexicana FMNH 13274 2.41 2.41 0.64 0.84 T. pumila FMNH 83610 2.35 2.35 0.76 T. pumila r 1 FMNH 83609 0.48 0.48 0.76 0.76 T. pumila FMNH 83607 0.48 0.68 T. pumila FMNH 83608 0.48 0.72 Molossus obscurus FMNH 55443 2.66 2.57 0.32 0.78 M. obscurus FMNH 55442 0.36 0.76 M. obscurus FMNH 55440 2.75 2.66 10 r .9 - Nyctinomus stehlini Mormopterus kalinowski • 13274 Molossus obscurus 55443* 55442* 83609 (I A r) 1069 (I* r) Tadarida mexicana T . . . . 83607/ Tadarida pumila \ J 47858(1) 7868(14 r) 7867 Tadarida cynocephala 83608 length 1 7 mm Graph 7. Length against width of the fenestra cochleae in Nyctinomus stehlini and some Recent Molossids. 78 SEGALL: INNER EAR IN FOSSIL BATS 79 Of even greater interest are the instances in which such similarity exists between a Recent and a fossil chiropter, as has been shown to be the case in Nycterobius gracilis and Myotis capaccinii. My studies show that their cochleae and fenestrae cochleae have similar shape and size (fig. 6, table 6, Fig. 7. Outline of the basal view of the cochlea and the postero-lateral view of the fenestra cochleae in A and A', Nyctinomus stehlini Basel S.G. 6240; B and B\ Mormopterus kalinowskii FMNH 19952; and C and C\ Tadarida cynocephala FMNH 7867. graph 6) and I have, therefore, assumed that the two taxa are closely related in spite of their other cranial differences. The phylogenetic implications of this are that the inner ear remains stable while the facial region evolved (see Revilliod, 1922, table, p. 135, for cranial and facial measurements). Apparently, the hearing mechanism adequately serves the Recent Myotis capaccinii about as well as it did the fossil Nyterobius gracilis, in spite of the changes in proportions of the facial structures necessary for food procurement and processing. A marked difference was noted in the dimensions of the cochlea and the fenestra cochleae between the Pseudorhinolophus (sp. indet.) which Schlosser ( 1 887, Art. 2) used for his illustration and the Ps. which Revilliod (1917) used for the description of Pseudorhinolophus schlosseri (table 1). This discrepancy indicates clearly that these materials belong to different species. ACKNOWLEDGMENTS This study was made possible by a grant from the Penrose Fund of the American Philosophical Society. I am obliged to the following in- stitutions for the loan of specimens: Naturhistorisches Museum, Basel, Switzerland; Geologisch-Palaontologische Abteilung des Hessischen Landesmuseum, Darmstadt, W. Germany; Muse'e d'Histoire Naturelle, Geneve, Switzerland; Universite de Lyon, Department des Sciences de la Terre, Lyon, France; Muse'e d'Histoire Naturelle, Montauban, France; Universite de Montpellier, Laboratoire de Pale'ontologie, Montpellier, France; Museum National d'Histoire Naturelle, Paris, France. I give my thanks for the great hospitality shown by Prof. R. Dehm, Dr. B. Engesser, 80 FIELDIANA: GEOLOGY, VOLUME 33 Doz. Fahlbusch, Dr. Heil, Dr. J. Hurzeler, Prof. R. Lavocat, Prof. J. P. Lehman, Dr. D. E. Russel, Dr. B. Sige', and Prof. H. Tobien. Thanks are due Dr. J. A. Hopson and Dr. W. Turnbull for reviewing the manuscript and to Dr. E. G. Wever for his technical advice. Mr. R. W. Roesener and Mr. Z. T. Jastrzebski did the artwork and Miss M. Alexander typed the manuscript. REFERENCES Dechasaux, C. 1958. Chiroptera. In Traite de Paleontologie, VI, 2d. Filhol, H. 1872. Recherches sur les mammife'res fossiles . . . Ann. Sci. Geol., T. III. 1876. Recherches sur les phosphorites du Quercy. Ann. Sci. Geol., T. VII. Gray, A. A. 1907. The labyrinth of animals. Vol. I. Churchill, London. Pictet, F. J., C. Gaudin and Ph. De La Harpe 1855. Memoir sur les animaux vertebres trouves dans le terrain siderothique du Canton de Vaud; Materiaux pour la paleontologie Suisse, Geneve. Pye, A. 1966. The structure of the cochlea in Chiroptera. I. Microchroptera, Emballonuroidea and Rhinolophidea. Jour. Morphol., 118, pp. 495-511. 1967. The structure of the cochlea in Chiroptera. III. Microchiroptera: Phyllostomatoidea. Jour. Morphol., 121, pp. 241-255. Revilliod, P. 1917. Memoires de la Societe pale'ontologique Suisse. Vol. 42. 1920. Me'moires de la Societe pale'ontologique suisse. Vol. 44. 1921-1925. Me'moires de la Societe' pale'ontologique Suisse. Vol. 45. Russel, D. E. and B. Sige 1970. Revision des chiropte'res lutetiens de Messel, Palaeovertebrata. Vol. 3, fasc. 4. Montpellier. Schlosser, M. 1887-1889. Die affen, lemuren, chiropte'ren etc. des europaischen Tertiars. BeitrSge zur Palaontologie Osterreich-Ungarns und des Orients. Vol. 6-7. Segall, W. 1970. Morphological parallelism of the bulla and auditory ossicles in some insectivores and marsupials. Fieldiana: Zool., 51, no. 15, pp. 169-205. 1971a. The auditory region (ossicles, sinuses) in gliding mammals and selected repre- sentatives of non-gliding genera. Fieldiana: Zool., 58, no. 5, pp. 27-59. 1971b. Auditory region in bats including Icaronycteris index. Fieldiana: Zool., 58, no. 9, pp. 103-108. SEGALL: INNER EAR IN FOSSIL BATS 81 SlGE, B. 1968. Les chiropte'res du Miocene infe'rieur de Bouzigues. Palaeovertebrata, vol. I, fasc. 3, Montpellier. Thompson, D. W. 1966. On growth and form, 2nd ed. Vol. 2. Cambridge University.