Silence or destruction by

microRNA p. 380

INSIGHTS

Conducting fibers that dont

fade in the stretch p. 382

PERSPECTIVES
PALEONTOLOGY

Downloaded from www.sciencemag.org on July 23, 2015

Four legs
too many?
A long-bodied fossil snake
retains fore- and hindlimbs
By By Susan Evans

Department of Cell and Developmental Biology, University

College London, London, UK. E-mail: s.e.evans@ucl.ac.uk

374

A four-limbed snake from the Cretaceous. Artists impression of Tetrapodophis, which retains four limbs, each with
five digits, in an elongated body with 160 precaudal vertebrae.
sciencemag.org SCIENCE

24 JULY 2015 VOL 349 ISSUE 6246

Published by AAAS

ILLUSTRATION: JULIUS CSTONYI

classic Gary Larson cartoon shows a

robed and bearded figure rolling out
clay strips, with the caption: God
makes the snake. Body elongation
was certainly fundamental in the evolution of snakes from lizards, as was
the shrinking and ultimately the loss of limb
pairs (limb reduction). However, informative
early fossils are rare, and many details of the
transition remain unresolved. A remarkable
fossil described on page 416 of this issue by
Martill et al. (1) brings fresh perspective to
the debate. The aptly named Tetrapodophis
combines a snakelike body with fore- and
hindlimbs bearing five well-developed digits
(see the illustration).
Snakelike bodies evolved several times
through geological history. Among amniotes
(reptiles, birds, and mammals), they occur
only in Squamata, the group comprising lizards and snakes. Within Squamata, however,
this body form has arisen independently
at least 26 times (2) (see the figure). Body
elongation is always correlated with limb
reduction (2), and the forelimbs are usually
lost first (Bipes and Bachia are rare exceptions). One explanation is that as the body
lengthens, coordination of limb movements
becomes increasingly difficult. Moreover, a
serpentine body moves most effectively by
lateral undulation, a movement in which
limbs can become a hindrance, especially
in narrow spaces. Researchers have identified a threshold body length at which limb

PHOTO: (A) D. M. MARTILL/UNIVERSITY OF PORTSMOUTH; ILLUSTRATION: (B) P. HUEY/SCIENCE

reduction begins, and no

sion domains, particularly
A
known squamate with more
those normally associated
than 70 precaudal (before-tail)
with the neck-trunk boundvertebrae retains four comary. They argued that the neck
plete limbs (2). Tetrapodophis
had been lost in snakes and
(1), with around 160 precauthat this loss disrupted the
dals, is therefore exceptional.
molecular signals required
Efforts to reconstruct the
for forelimb positioning and
10 mm
evolutionary stages in the
outgrowth. However, in anTetrapodophis
B
snake body plan are hamother snake, Pantherophis,
pered by a lack of consensus
the Hox expression domains,
Scincidae
Gymnophthalmidae
on snake relationships and analthough expanded and withLerista
cestral lifestyle. Analyses using
out sharp boundaries, retain
molecular data group snakes
a regionalized pattern compaBachia
with terrestrial lizards like
rable to that of lizards with a
Acontias
iguanas and Komodo dragons
distinct neck (9, 11). A parallel
(Iguania and Anguimorpha)
study of vertebral anatomy
Anguidae
(3) and generally hypothesize
across a wide range of snakes
Lialis
a burrowing or semiburrow(12) revealed a similar regioning ancestry (13). However,
alized pattern, implying that
Ophisaurus
some analyses that include
snakes have a neck of 10 to 12
Gekkota
anatomical characters place
segments.
them with extinct Cretaceous
Like that of a lizard, the
Pantherophis
Dibamus
(~100 to 66 million years ago)
vertebral column of TetraBipes
marine lizards, the mosasaurs
podophis has distinct regions,
Dibamidae
(3, 4). This has prompted the
including 10 to 11 short-ribbed
suggestion of a marine swimneck vertebrae adjacent to
Serpentes
Amphisbaena
ming ancestry for snakes (4).
the tiny forelimbs. This neck
Molecular divergence eslength is within the range of
Amphisbaenia
Hassiophis
timates date snake origins
some generalized terrestrial
to the Jurassic (~150 million
Limbs or no limbs. (A) Martill et al. report the discovery of a four-limbed snake,
lizards and matches that proyears ago) (5), but the earliest
Tetrapodophis amplectis, from the Cretaceous. (B) Schematic showing independent
posed by the developmental
uncontested fossils are isodevelopment of the long bodied, limb-reduced body plan among squamates (not to scale).
(9, 11) and anatomical (12)
lated vertebrae from the midstudies. Thus, as in long-bodCretaceous (~113 million years ago) of North
lion years old). Martill et al. (1) place it on
ied lizards, elongation of the snake skeleton
America (5). These vertebrae come from
the stem of the snake evolutionary tree, beoccurred in the trunk region and not the
terrestrial deposits but are otherwise fairly
low Najash and close to another early terresneck. Moreover, if Tetrapodophis is correctly
uninformative. More instructive are several
trial snake, the North American Coniophis,
interpreted as a stem-snake, that elongation
articulated skeletons or partial skeletons
represented by vertebrae and attributed jaw
preceded loss of the forelimbs.
from slightly younger (~100 million-year-old)
elements.
Love them or loathe them, snakes have
deposits. The largest set consists of several
Whereas fossils can yield information on
long fascinated humans. The combined efrelated marine snakes from the Middle East,
the sequence of anatomical changes involved
forts of paleontology and developmental
North Africa, and southern Europe. These
in any major transition, developmental biolbiology have gone some way toward unravfossil snakes have 140 to 155 precaudal verogy helps to explain how these changes oceling the early history of snakes, but many
tebrae and a short tail. They show no trace
curred. Evolution of the snake body form
questions remain as to their origins, relaof forelimbs or shoulder girdle but do have
combined axial elongation, limb loss, and
tionships, character evolution, and ancessmall hindlimbs; only one [Haasiophis (6)]
reduced regionalization (8, 9). Whether and
tral lifestyle. Resolution of these questions
preserves digits. The relationships of these
how these components are linked developdepends, ultimately, on the recovery of furlimbed marine snakes remain controvermentally remains uncertain. In all vertebrate
ther fossils and their thorough and objecsial, but many analyses (1, 3, 6, 7) nest them
embryos, individual vertebrae develop from
tive analysis.
among modern snakes, rather than nearer
segments (somites) that form at regular
REFERENCES
the base of the snake evolutionary tree. This
intervals. To increase vertebral numbers,
1. D. M. Martill et al., Science 349, 416 (2015).
implies either that hindlimbs were reduced
somite formation must either continue for
2. M. C. Brandley, J. P. Huelsenbeck, J. J. Wiens, Evolution 62,
more than once within snakes, or that the
longer or occur at a faster rate. Snakes use
2042 (2008).
3. T. W. Reeder et al., PLOS ONE 10, e0118199 (2015).
limbs redeveloped in some lineages (6).
both strategies (8). Individual vertebrae then
4. M. W. Caldwell, Zool. J. Linn. Soc. 125, 115 (1999).
A second set of early fossil snakes comes
acquire positional identity along the body
5. J. J. Head, Palaeontol. Electronica 18, 1 (2015).
from terrestrial deposits in South America.
axis through the overlapping expression do6. E. Tchernov, O. Rieppel, H. Zaher, M. J. Polcyn, L. L. Jacobs,
Science 287, 2010 (2000).
The most complete, 95-million-year-old Namains of Hox genes. In a typical tetrapod, the
7. S. Apestegua, H. Zaher, Nature 440, 1037 (2006).
jash (7), resembles the fossil marine snakes
boundaries between major vertebral regions
8. C. Gomez et al., Nature 454, 335 (2008).
in having small hind legs without preserved
(such as the neck and the trunk) coincide
9. J. M. Woltering et al., Dev. Biol. 332, 82 (2009).
10. M. J. Cohn, C. Tickle, Nature 399, 474 (1999).
digits but is more primitive (1, 7). Tetrapodowith Hox gene expression boundaries.
11. N. Di-Po et al., Nature 464, 99 (2010).
phis is also from South America, and from a
In a pioneering study of Python devel12. J. J. Head, P. D. Polly, Nature 520, 86 (2015).
deposit that yields a mix of freshwater and
opment, Cohn and Tickle (10) reported a
10.1126/science.aac5672
terrestrial species, but it is older (~113 milmarked expansion of the typical Hox expresSCIENCE sciencemag.org

24 JULY 2015 VOL 349 ISSUE 6246

Published by AAAS

375

Four legs too many?

Susan Evans
Science 349, 374 (2015);
DOI: 10.1126/science.aac5672

This copy is for your personal, non-commercial use only.

If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.
Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.

Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/349/6246/374.full.html
A list of selected additional articles on the Science Web sites related to this article can be
found at:
http://www.sciencemag.org/content/349/6246/374.full.html#related
This article cites 12 articles, 2 of which can be accessed free:
http://www.sciencemag.org/content/349/6246/374.full.html#ref-list-1
This article appears in the following subject collections:
Geochemistry, Geophysics
http://www.sciencemag.org/cgi/collection/geochem_phys

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2015 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.

Downloaded from www.sciencemag.org on July 23, 2015

The following resources related to this article are available online at

www.sciencemag.org (this information is current as of July 23, 2015 ):