skeletons, from the Holzmaden area of southern Germany,
confirmed that the tailbend was natural, and that the
down-turned portion of the vertebral column supported
the lower lobe of a broad, crescentic tail. The skin impressions
also revealed that the paired fins were broad-based,
and that there was a dorsal fin without bony support. Such
deep-tailed, broad-finned animals were clearly not adapted
for hauling themselves up on land.
The quarries in the Posidonienschiefer (Upper Lias;
Lower Jurassic: Toarcian) of Holzmaden and vicinity in
Württemberg, Germany, have been worked for slate for
several hundred years, and ichthyosaurian remains were
discovered there over two centuries ago (Ziegler, 1986).
However, as in England, the significance of these specimens
was not realized until much later. The Holzmaden
quarries, being considerably larger and more productive
than their English counterparts, yielded far more skeletons,
and Germany soon surpassed England for its ichthyosaurian
discoveries. Research attention tended to shift
from England, and the end of the nineteenth and beginning
of the twentieth century saw a proliferation of German
publications. Major contributors included E berhard
Fraas, of the Royal Natural History Museum in Stuttgart
(the precursor of today’s Staatliches Museum für Natur-
kunde Stuttgart), and F riedrich von Huene, of the University
of Tübingen. F raas (1891) and H uene (1922) represent
particularly important monographs on ichthyosaurs. Meanwhile,
in England, Andrews (1910) made a major contribution
to our understanding of Late Jurassic forms with his
monographic study on Ophthalmosaurus. This was based
upon the extensive collection of material from the clay pits
in the Oxford Clay of Peterborough and vicinity, which are
still being worked today for clay, used in the manufacture
of bricks. Although the specimens are well preserved and
numerous, they are largely dissociated, and the only complete
skeleton, on display in the Natural History Museum,
London, is a composite of two individuals. Andrews’s
(1910) contribution to our knowledge of Ophthalmosaurus
has subsequently been surpassed by the outstanding work
of Kirton (1983). One of the most important contributions
to our knowledge of the cranial anatomy was made by
Sollas (1916) with his treatment of Ichthyosaurus, studied
in serial section.
Much of the very extensive literature on ichthyosaurs
deals with their morphology and taxonomy. However,
there have been many studies on their affinities, and upon
the unresolved question of their ancestry. Callaway (1997)
provided an excellent historical overview of this research.
By far the greatest number of ichthyosaurs that have
been found are from post-Triassic strata, mostly from the
Early Jurassic of England and Germany. The quarries of
Holzmaden and vicinity have been the most productive,
and it is estimated that they have yielded about three
thousand skeletons (McGowan, 1991). Thus Early Jurassic
species are very well known, and most of our knowledge
of ichthyosaurian anatomy pertains to these “typical”
forms. Triassic ichthyosaurs, in contrast, are rather poorly
known, due both to the generally fragmentary nature of
the material and to its relative scarcity. The notable exception
is the Middle Triassic locality of Monte San Giorgio in
southern Switzerland, on the border with Italy. This rich
site has yielded large numbers of complete, excellently
preserved skeletons of Mixosaurus (Bürgin et al., 1989).
This small ichthyosaur, first described by Bassani in 1886,
has pentadactyl limbs with only modest hyperphalangy,
and a barely developed tail flexure, making it a seemingly
ideal basal form.
The first Triassic ichthyosaurs were found in southern
Germany, several decades before the discovery of the famous
locality at the Monte San Giorgio. Material has also
been found in Spitzbergen and these fossils are Early Triassic
in age, and represent some of the geologically oldest
ichthyosaurs known to date. Unfortunately, like most other
Triassic material, these specimens are mostly fragmentary.
Other important Triassic ichthyosaur-producing localities
are in North America, China and Japan.
Merriam’s (1908) monograph was the first attempt to
bring order to this wide and largely fragmentary array of
Triassic material, and this was followed by H uene (1916).
Callaway (1989) added to this work, with a major contribution
on the systematics and phylogeny of Triassic ichthyosaurs.
Motani (1996,1997A-C, 1998,1999B), in a series
of papers, shed new light on Triassic ichthyosaurs, correcting
many misinterpretations of the past. This prepared the
way for the present understanding of ichthyosaurian relationships
(Motani et al., 1998, Motani, 1999C).
The first Cretaceous ichthyosaurian material was described
by Carter (1846A, 1846B) from the Lower Cretaceous
of England. Most of these specimens are fragmentary
- mainly isolated teeth, jaw fragments, vertebral centra,
and phalanges. However, more complete material has
since been found in North America and Australia. Our
knowledge of the last ichthyosaurs, represented by the
single genus Platypterygius, is accordingly quite good.
Ichthyosaurs were probably worldwide in their distribution,
because of their well-developed swimming abilities
and marine habitat. Their present geographical distribution
therefore probably just reflects the localized nature
of good fossiliferous exposures (McGowan, 1978). Their
geological range extends from the Early Triassic (Oleneki-
an) to the Late Cretaceous (Cenomanian), but we only see
occasional glimpses of this long history - a few still frames
from a continuous sequence. Most of these “snapshots”
date from the Early Jurassic, and this has tended to color
our perceptions of their anatomy and interrelationships.
Kiprijanoff (1881), for example, recognized two groups
among Liassic ichthyosaurs, based on the possession of
broad or narrow forefins. These two fin types, subsequently
referred to as latipinnate and longipinnate, were thought
to represent a post-Triassic dichotomy of the group, which
was later extended to include Triassic forms (McGowan,
1972A). However, as more non-Liassic material was examined,
it became clear that the dichotomy was apparent
rather than real (McGowan, 1976), and most specialists
have now abandoned this terminology. The Jurassic bias of
the fossil record has also been a contributing factor to the
unsuccessful attempts to resolve the vexed question of
ichthyosaurian origins. The key to this problem is a sound
understanding of Early Triassic ichthyosaurs, an area in
which RM has recently made important contributions.
From their earliest appearance in the fossil record ichthyosaurs
were adapted for aquatic life, with paired fins
instead of walking limbs. There is also evidence that the
earliest ichthyosaurs had some kind of caudal fin, though
without a tailbend (Motani et al., 1996). These earliest
ichthyosaurs were long-bodied, and probably used an
anguiUiform swimming mechanism (Motani et al., 1996).
The tailbend, which is present in all post-Triassic ichthyosaurs
(contra Riess, 1986; see McGowan, 1990A), was certainly
present by Late Triassic times, as exemplified by
Shonisaurus from North America (Kosch, 1990; McGowan
& Motani, 1999), and by several geologically older ichthyosaurs
from China, including Guanlingsaurus and Qian-
ichthyosaurus (Yin et al., 2000). A tailbend has also been
reported for the Middle Triassic Cymbospondylus (Hogler
& Kosch, 1993).
Post-Triassic ichthyosaurs evolved a thunniform body
shape, like that of extant scombroid fishes and cetaceans,
with a crescentic tail and streamlined body. The caudal fin
of ichthyosaurs has generally been interpreted as functioning
like the reversed heterocercal tail of sharks (McGowan,
1973A; Taylor, 1987). This type of tail was believed to
produce a downward force, counteracting the positive
buoyancy of the air-breathing ichthyosaur, thereby initiating
diving. Such a scenario is based on the assumption that
shark tails generate an asymmetric force, attributable to
the greater flexibility of the skeletally unsupported ventral
lobe. This assumption is true for some sharks, such as the
dogfish (Scyliorhinus caniculus) and tope (Galeorhinus ga-
leus), upon which the early investigations of the function of
the heterocercal tail were conducted (Alexander, 1965).
However, the unsupported ventral lobe is actually stiffer
than the dorsal lobe in many other species, including the
hammerhead shark (Sphyrna lewini), the shortfin mako
(Isurus oxyrinchus), the blue shark (Prionace glauca), and
several species of requiem sharks (Carcharhinus; McGowan,
1992). A more likely scenario for ichthyosaurs is that
the entire tail was stiff, and that it generated only horizontal
thrust, with no vertical component (McGowan, 1992).
Diving might have been initiated in a surface-swimming
individual by flexing the body to depress the head. Changes
in horizontal swimming level were probably effected
using the pectoral fins as inclined planes (McGowan,
1992).
Ichthyosaurs, in contrast to plesiosaurs, appear to have
been laterally rather than dorsoventrally compressed,
though some, like Ophthalmosaurus, seem to have been
remarkably barrel-chested. Consequently, when ichthyosaurs
died and sank to the seabed, their bodies tended to
come to rest on their sides, rather than on their backs or
bellies. The laterally compressed tail probably helped impose
this post-mortem orientation. Most ichthyosaurian
skeletons are therefore found laterally compressed, with a
much lower incidence of dorsoventrally compressed ones.
Post-Triassic ichthyosaurs all have conical teeth, the upper
ones intermeshing with the lower ones. Such a dentition is
characteristic of animals that feed on fishes and cephalo-
pods, such as extant toothed whales. This conclusion is
supported by evidence from gut contents, and from copro-
lites that are purportedly ichthyosaurian. The gut contents
predominantly comprise cephalopod arm hooklets, with
few instances of fish scales, whereas coprolites have only
fish scales (Pollard, 1968). This suggests that while fish
scales were eliminated (or digested) from the gut in the
usual way, the cephalopod hooklets were retained in the
stomach to prevent damaging the lining of intestine. The
hooklets may then have been periodically vomited from
the stomach, as in the present-day sperm whale.
Massare (1987) gave a detailed analysis of tooth types
among marine predators, recognizing seven overlapping
feeding guilds. These ranged from the “piercing guild” of
predators, with slender pointed teeth for piercing soft prey,
to the “cut guild”, with robust teeth for seizing large marine
vertebrates. These guilds are exemplified by Lepto-
nectes tenuirostris and Temnodontosaurus eurycephalus, respectively.
Whereas most ichthyosaurs have conical teeth,
some species are completely or effectively edentulous. In
Stenopterygius quadriscissus, for example, young individuals
have well-developed teeth, whereas most mature individuals
have minute teeth, or no teeth at all. Although
teeth occur in Ophthalmosaurus they appear to have been
very loosely attached to the jaws, and are usually lost
during preservation (Kirton, 1983). Mature individuals
may have been edentulous. Triassic ichthyosaurs also have
conical teeth, at least anteriorly, but in many species they
become low-crowned and blunt posteriorly.
Ichthyosaurs typically have large eyes, as shown by the
large diameter of the orbit and scleral ring. Indeed, Motani
(1999D) has demonstrated that ichthyosaurs have relatively
larger eyes than any other vertebrates of similar body
size. It is therefore likely that vision was their predominant
sense. This is supported by endocranial evidence. Although
scant, this evidence indicates that the architecture of the
brain was dominated by the large size of the optic lobes
(McGowan, 1973B). The olfactory lobes appear to have
been only modestly developed, suggesting that olfaction
may not have been as significant in ichthyosaurs as it is in
many extant diapsids.
A prerequisite for directional hearing underwater is
that the two inner ears be acoustically isolated. This is