ArticlePDF Available

An evaluation of techniques for measurements of snake length

Authors:

Abstract and Figures

The efficacy of two measurement techniques was evaluated in a laboratory setting for user variation and practicality. A total of 59 snakes (15 Lampropeltis getula floridana, 16 Pantherophis guttatus, 12 Epicrates cenchria maurus, and 16 Thamnophis sauritus) were measured using traditional soft-tape measurements paired with restraining tubes. The second measurement method evaluated snake length using the digital imaging software Snakemeasurer© by taking a photo parallel to the surface the snake was resting on with a known length object in the photo for measurement reference. Each snake was measured by two designated measurers (one experienced and one recently-trained) using both measurement techniques. Each researcher and technique produced similar measurements. Digital measurements were not significantly different between measurers while soft-tape measures varied with species.
Content may be subject to copyright.
Collinsorum 2(1/2) March/June 2013 20
www.cnah.org/khs/
INTRODUCTION
Varying types of size measurements are used
in studies on snakes and can provide valuable
data (Fitch, 1987). The typical length mea-
sure reported for snakes is the snout-to-vent
length (SVL) as it is seen as the most impor-
tant length measurement (Fitch, 1987), but
this has not always been the standard (Sei-
gel and Ford, 1988). Seigel and Ford (1988)
reported the three most common measure-
ments of snakes to be SVL, total length (TL),
and mass. While SVL is the standard length
measurement used for snakes today, TL is still
occasionally used (Penning and Cairns, 2012).
Typically SVL measurements are recorded us-
ing the methodologies of Fitch (1987) due to its
time and cost effectiveness in the field. A soft
or hard tape measure is used and the snake is
gently stretched along the tape length. Other
methods of measurement that have been used
include restraining tubes (Fitch, 1987), press-
boxes (Quinn and Jones, 1974; Bertram and
Larsen, 2004), anesthesia (Blouin-Demers,
2003), and digital imaging software (Measey
et. al., 2002). All of the above mentioned mea-
surement techniques are prone to three types
of error: single user-single measurement tech-
nique variation; multiple user-single measure-
ment technique variation; and multiple mea-
surement technique variation (Setser, 2007).
Highly precise and repeatable measurements
are becoming increasingly important, espe-
cially with the newly discovered bidirectional
growth phenomenon reported in marine igua-
nas (Wikelski and Thom, 2000). This phenom-
enon was investigated in snakes using data
from a long-term field study on Liasis fuscus,
and based on recapture data, 6.42% of the
recaptures reported shrinking lengths (Mad-
sen and Shine, 2001). The same phenomenon
was reported in Blouin-Demers et al. (2002)
for Pantherophis obsoletus. Both of these ob-
servations were reported to be measurement
error and not actual shrinking. Contrastingly,
measurement error has not always been con-
sidered to cause false results in experimental
evaluations (Merila and Bjorklund, 1995).
Several investigators have evaluated the ac-
curacy and precision of length measures of
elongate vertebrates using various measure-
ment techniques and varying levels of support
exist for every measurement type. Quinn and
AN EVALUATION OF TECHNIQUES FOR
MEASUREMENTS OF SNAKE LENGTH
David Penning1,2,3, Eva Gann1, William Thomas1, Tyler Carlson1, Jennifer
Mittelhauser1, Leslie Bilbrey1, Stefan Cairns1
1Department of Biology and Earth Science, University of Central Missouri,
Warrensburg, Missouri, USA
2Department of Biology, University of Louisiana at Lafayette, Lafayette,
Louisiana, USA
3CORRESPONDENCE: e-mail, davidapenning@gmail.com
Abstract: The ef cacy of two measurement techniques was evaluated in a laboratory setting for
user variation and practicality. A total of 59 snakes (15 Lampropeltis getula oridana, 16 Panthero-
phis guttatus, 12 Epicrates cenchria maurus, and 16 Thamnophis sauritus) were measured using
traditional soft-tape measurements paired with restraining tubes. The second measurement method
evaluated snake length using the digital imaging software Snakemeasurer© by taking a photo paral-
lel to the surface the snake was resting on with a known length object in the photo for measurement
reference. Each snake was measured by two designated measurers (one experienced and one
recently-trained) using both measurement techniques. Each researcher and technique produced
similar measurements. Digital measurements were not signi cantly different between measurers
while soft-tape measures varied with species.
Key Words: digital measure, snake length, Snakemeasurer©, straight line length
Collinsorum 2(1/2) March/June 2013 21
www.cnah.org/khs/
Jones (1974) found the squeeze-box tech-
nique to produce reliable measurements with
little variation, while Setser (2007) found the
same technique to produce less reliable results
than anesthetizing snakes prior to straight
measurement. Bertram and Larsen (2004) re-
ported significantly larger measurements to be
produced by straight-line measurements when
compared to squeeze-box measurements and
recommended one tracing in a squeeze-box
measured three times to best balance han-
dling time and accuracy. Cross (2000) evalu-
ated squeeze-box and anesthetized measure-
ment techniques and found <1 cm variation
between measurements. Measey et al. (2002)
compared fixed-ruler measurements to flexible
measurements using digital imaging software
(Wilcox et al., 1997) and found both measures
to be reliable with fixed-rulers producing sig-
nificantly longer measurements. Blouin-De-
mers (2003) found all three of the investigated
measurement forms (soft-tape while awake,
soft-tape while anesthetized, and solid-tape
while anesthetized) to be effective and con-
cluded with the recommendation for measur-
ing snakes under anesthesia using a solid ruler.
Measurements using soft-tape while the snake
is awake is a reasonable alternative, a conclu-
sion also supported by Setser (2007).
Using different measurement techniques not
only come at varying economical costs (Setser,
2007), but also have potential ecological im-
pacts (Fitch, 1987). The process of measuring
snakes awake with a soft-tape in the field has
been shown to have negative growth impacts
for several weeks after capture in Crotalus viri-
dis (Fitch, 1949). Using anesthesia to measure
snakes brings the greatest risk (Fitch, 1987)
and has been cautioned against its use if only
measurement data are to be taken (Measey
et al., 2002). Hand measurements have the
drawback of being a “one off” measurement
with no capability of being checked by an alter-
nate authority (Measey et. al., 2002). Errors in
snake measurements are an important prob-
lem (Fitch, 1987) that needs to be addressed
with minimally invasive handling techniques
in mind (sensu Fellers et al., 1994). Measure-
ment error of SVLs have been found to be so
variable that its use has occasionally been
abandoned (Houston and Shine, 1994), an
unfortunate outcome that may be prevented
with new software. Snakemeasurer© is digital
imaging software designed to measure snakes
based on single-frame images with a known
measure within the image. It is freely avail-
able (http://serpwidgets.com/main/measure)
and has been used in prior publications (Pen-
Figure 1. An overhead image of all four snake species [Lampropeltis getula oridana (A); Epicrates cenchria maurus (B),
Pantherophis guttatus (C), and Thamnophis sauritus (D)] using 0.4 cm graph paper as the measurement reference.
AB
CD
Collinsorum 2(1/2) March/June 2013 22
www.cnah.org/khs/
ning and Cairns, 2012). The main objective of
our study is to evaluate the effectiveness of
Snakemeasurer© for measuring snake length
and how it compares to the classic soft-tape
measuring technique.
METHODS
Snakes used in this study came from vari-
ous growth and behavioral studies being per-
formed at the University of Central Missouri
during spring 2012. A total of 59 snakes were
used for the evaluation of measurement tech-
niques (15 Lampropeltis getula floridana, 16
Pantherophis guttatus, 12 Epicrates cenchria
maurus, and 16 Thamnophis sauritus). Man-
ual measurements involved the use of a soft-
tape measure and various restraining tubes.
Snakes were placed in a restraining tube just
large enough to accommodate them and were
then measured for total length. Total length
was used due to the overhead imaging pro-
cedures used with the software program. As
we are not making any ecological implications
based on our length data, total length serves
the identical purpose for comparing measures.
All snake total lengths were measured twice
using soft-tape (once by an experienced re-
searcher and once by a recently-trained re-
searcher) to the nearest 0.1 cm. Snakes were
then placed in a 37.8 L aquarium lined with
0.4 cm graph paper where they were free to
orient themselves. Snakes were photographed
from above with the camera on a parallel plain
to the aquarium bottom (Penning and Cairns,
2012). Occasionally snakes were moved gen-
tly with a snake hook until an acceptable body
position was available for imaging (no overlap
or raised body sections). The snake needed to
be flat with no portion of the body elevated
from the substrate because the imaging soft-
ware measures only in two dimensions. Both
the tip of the snout and tip of the tail needed
to be easily viewed within the image. Images
were then run through Snakemeasurer© for
straight-line length by both researchers using
the graph paper as the known measurement
length (Fig 1). The experienced researcher re-
mained the same for all measurements but the
recently-trained researchers changed between
snake species with the exception of L. g. flori-
dana and P. guttatus.
All statistical analyses were performed on
MS EXCEL and MINITAB. Paired t-tests (two-
tailed) were performed between measurers of
each technique for each species due to the re-
peated measure design of data collection (Mc-
Donald, 2008), as well as its previous use in
similar investigations (Measey et al., 2002; Bl-
ouin-Demers, 2003). Variables were normally
distributed and all tests were considered sig-
nificant at p<0.05.
RESULTS
Implementation
For the experienced user the time to com-
plete the entire procedure was, on average,
faster for soft-tape measurements (ca. 2 min).
Overall time spent handling snakes was much
faster for the digital imaging (<1 min), but
required more time for image transfer, down-
loading, and processing (ca. 3 min).
Reliability
A total of 236 measurements were taken from
59 snakes. For L. g. floridana, mean straight
line lengths varied only by 1 mm between digi-
tal measurers and were not significantly dif-
ferent (paired t-test (15) = -0.453, p>0.05),
while soft-tape measurements were signifi-
cantly different with the average varying by 12
mm (paired t-test (15) = -5.467, p<0.05). For
P. guttatus, mean straight line lengths varied
by 5 mm between digital measurers and 1 mm
between soft-tape measurers and both were
not significantly different between measur-
ers (paired t-test (14) = -2.08, p>0.05 and
paired t-test (14) = -0.4483, p>0.05 respec-
tively). For E. c. maurus, mean straight line
Species Digital Measure Soft Tape Measure Paired t-test
L. g. floridana 511±8.7 mm 478±8.5 mm t(15) = 15.34, p<0.05
P. guttatus 625±9.14 mm 590±8.32 mm t(15) = 16.27, p<0.05
E. c. maurus 963±11.08 mm 885±10.23 mm t(11) = 25.22, p<0.05
T. sauritus 609±4.78 mm 591±5.09 mm t(15) = 14.44, p<0.05
Table 1. Combined measurement (mean±SD) results from four snake species (a total of 236 measure-
ments). Digital snake measuring using Snakemeasurer© is represented by “Digital Measure”, hand mea-
surements using restraining tubes and a soft ruler is represented by “Soft-Tape Measure”.
Collinsorum 2(1/2) March/June 2013 23
www.cnah.org/khs/
lengths were not significantly different among
digital measurers (paired t-test (11) = 2.128,
p>0.05) while soft-tape measurers were
(paired t-test (11) = -12.749, p<0.05). For
T. sauritus, mean straight line lengths varied
by 6 mm and were not significantly different
between digital measurers (paired t-test (15)
= 1.777, p>0.05) while mean straight line
length was significantly different between soft-
tape measurers (paired t-test (15) = -7.590,
p<0.05). All combined digital measures were
significantly larger than combined soft-tape
measures (Table 1).
DISCUSSION
Our results are similar to that of Measey et
al. (2002) in that one measure consistently
produced significantly longer measurements
than the other but differ in the longer mea-
sure reported. In our study, Snakemeasurer©
software reliably produced longer measure-
ments compared to the soft-tape measure-
ments, markedly so in E. c. maurus. This spe-
cies is a notorious resister to manual restraint
and the forces at which it can do so have been
quantified (Lourdais et al., 2005). All digital
measurements were similar when compared
between researchers while the soft-tape mea-
sures were significantly different between re-
searchers in all but the P. guttatus compari-
son. This supports the notion that measurer
variation is higher in hands-on manipulations
of snakes and less when using digital imaging
software.
Functionally, Snakemeasurer© digital imag-
ing software is similar to anesthetized mea-
sures. The snake is not active and time can
be taken to measure length carefully with no
struggling, resistance, or spinal flexion. Bl-
ouin-Demers (2003) found the greatest varia-
tion in measurements to be active snakes and
the most precise measurements to come from
anesthetized snakes. These results support
this trend by showing no significant difference
between researchers using Snakemeasurer©
digital imaging software. When compared be-
tween researchers, soft-tape measurements
were significantly different in three of the fours
species we investigated.
Palmeirim (1998) reported skull measure-
ment variation between different measurers
to be 30.7%, almost twice the value of intra-
measurer variation. Ecological studies can
span several investigators and measurer bias
and error has the ability to impact measure-
ment recordings (Madsen and Shine, 2001;
Blouin-Demers et al., 2002) even to the ex-
tent of rejecting length as a measure in gen-
eral (Houston and Shine, 1994). With powerful
musculature controlling the snake’s ability to
move and in some cases slightly elongate or
compress during handling, gentle stretching
procedures used in many measurement inves-
tigations could have and have yielded variable
results from experimenter manipulation (Hous-
ton and Shine, 1994). Coupling measurer and
measurement technique variation could have a
large impact on the measurements reported.
The use of a measuring technique that elimi-
nates as many potential errors balanced with
a lessened post-measurement impact on the
organism of investigation would be the ideal
technique.
In this study, Snakemeasurer© digital imag-
ing software showed the least variation be-
tween measurers compared to soft-tape mea-
suring and has added benefits that no other
measuring technique can provide. Inter-mea-
surer error can be completely eliminated with
this technique even if the length measurer was
not the field investigator. Additionally, with the
ability to easily store images it is possible to
allow additional researchers access to the data
allowing for the further elimination of inter-
measurer error even in large compiled data
sets.
Measey et al. (2002) listed many of the ben-
efits of using digital imaging software, but at
the time, warned of its expense and techni-
cal complexity. In today’s market, digital cam-
eras can be purchased at reasonable prices;
the Snakemeasurer© digital imaging software
is freely available, and the newly trained re-
searchers in this study had no difficulty using
the software. It is a much less expensive alter-
native to anesthesia and in this study had none
of its reported drawbacks (Setser, 2007). Han-
dling time per snake was reduced, although
there was a slight increase in total measuring
time using the digital software. The expense of
added time is a cost to the researcher with the
benefit of reduced handling to the snake. The
simplest and most cost-effective measuring
technique will likely always be soft tape mea-
suring but even this fairly noninvasive tech-
nique has been shown to have a negative im-
pact on snake growth (Fitch, 1949). The most
precise measurement technique (anesthesia)
is the most expensive. Its use requires the
most time for the researcher as well as han-
Collinsorum 2(1/2) March/June 2013 24
www.cnah.org/khs/
dling time for the snake and can prove to be
fatal in some cases (Setser, 2007). When de-
signing future experiments, researchers should
design their experiments with all of the above
benefits and drawbacks in mind.
ACKNOWLEDGEMENTS
We would like to thank the Department of Bi-
ology and Earth Science faculty and staff, spe-
cifically Scott Lankford, Joseph Ely, and Dan-
iel Metcalf for their help and guidance. This
research was partially funded by The Honors
College grant of Eva Gann and undergraduate
research grants of William Thomas and Tyler
Carlson from the University of Central Missouri.
We would also like to thank Mark Miles of Miles
of Exotics in Kansas City, Missouri for his signif-
icant financial contribution that allowed this re-
search to happen. This research was conducted
under IACUC protocol #10-3212.
LITERATURE CITED
Bertram, N., and K.W. Larsen. 2004. Putting
the squeeze on venomous snakes: accuracy
and precision of length measurements taken
with the “squeeze box”. Herpetological Re-
view 35: 235-238
Blouin-Demers, G., Prior, K.A., and P.J. Weath-
erhead. 2002. Comparative demography of
black rat snakes (Elaphe obsoleta) in Ontario
and Maryland. Journal of Zoology 256: 1-10
Blouin-Demers, G. 2003. Precision and accura-
cy of body-size measurements in a constrict-
ing, large-bodied snake (Elaphe obsoleta).
Herpetological Review 34: 320-323
Cross, C.L. 2000. A new design for a light-
weight squeeze box for snake field studies.
Herpetological Review 31: 34
Fellers, G.M., Drost, C.A., and W.R. Heyer.
1994. Handling live amphibians. In Heyer,
W. R., Donnely, M.A., and R. W. McDiarmid
(eds.), Measuring and Monitoring Biological
Diversity, Standard Methods for Amphibians.
Smithsonian Institutional, Washington D.C.
Fitch, H.S. 1949. Study of snake populations in
central California. American Midland Natural-
ist 41: 513-579
Fitch, H.S. 1987. Collecting and life-history
techniques. In Seigel, R.A., Collins, J.T., and
S.S. Novak (eds.), Snakes: Ecology and Evo-
lutionary Biology, pp 143-164. MacMillan
Publishing, New York
Houston, D., and R. Shine. 1994. Low growth
rates and delayed maturation in Arafura file-
snakes (Serpentes: Acrochordidae) in tropi-
cal Australia. Copeia 1994: 726-731
Lourdais, O., Brischoux, F., and L. Barantin.
2005. How to assess musculature and perfor-
mance in a constricting snake? A case study
in the Columbian rainbow boa (Epicrates cen-
chria maurus). Journal of Zoology 265: 43-51
Madsen, T., and R. Shine. 2001. Do Snakes
Shrink? Oikos 92:187-188
McDonald, J.H. 2008. Handbook of Biological
Statistics. Sparky House Publishing, Maryland
Measy, G.J., Silva, J.B., and M. Di-Bernardo.
2002. Testing for repeatability in measure-
ments of length and mass in Chthonerpeton
indistinctum (Amphibia: Gymnophiona), in-
cluding a novel method of calculating total
length of live caecilians. Herpetological Re-
view 33
Merila, J., and M. Biorklund. 1995. Fluctuating
asymmetry and measurement error. System-
atic Biology 44: 97-101
Palmeirim, J.M. 1998. Analysis of skull mea-
surements and measurers: Can we use data
obtained by various observers? Journal of
Mammalogy 79: 1021-1028
Penning, D.A., and S. Cairns. 2012. Growth
rates of neonate red cornsnakes, Panthero-
phis guttatus (Colubridae), when fed in mu-
tually exclusive mass-ratio feeding catego-
ries. Herpetological Review 43: 605-607
Quinn, H., and J.P. Jones. 1974. Squeeze box
technique for measuring snakes. Herpetolog-
ical Review 5:35
Seigel, R.A., and N.B. Ford. 1998. A plea for
standardization of body size measurements
in studies of snake ecology. Herpetological
Review 19: 9-10
Setser, K. 2007. Use of Anesthesia Increases
Precision of Snake Length Measurements.
Herpetological Review 38: 409-411
Wikelski, M., and C. Thom. 2000. Marine Igua-
nas Shrink to Survive El Nino. Nature 403:
37-38.
Wilcox, C.D., Dove, S.B., McDavid, W.D., and
D.B. Greer. 1997. UTHSCSA ImageTool Ver-
sion 2.00 http://ddsdx.uthscsa.edu/dig/
... The camera system employed by Yackel Adams et al. (2019) reliably recorded live-lure contact rates but was not optimized for measuring snake size. Snake length can be measured from photographs when the entire length of the snake is in the same frame along with a size standard (Penning et al. 2013). However, in the field, it is extremely difficult to get reliable images of a snake's entire body on a controlled plane at a fixed or known distance from the camera or with adequate size standards, particularly for an arboreal snake. ...
... Head length alone is a relatively precise predictor of brown treesnake length, explaining 95.9% of the variation in SVL after removing outlier measurements. Given the challenge of measuring a prehensile and uncooperative snake over a flexible tape, and interobserver differences in the force with which snakes are stretched (Penning et al. 2013, Cundall et al. 2016, much of the remaining 4.1% of unexplained variation could be due to error in SVL measurement. It may be that brown treesnake head length is a more precise predictor of body length than direct measurement of SVL with a flexible tape, given the less difficult task of measuring a relatively small, rigid, and easily restrained part of the anatomy with calipers. ...
Article
Full-text available
As in many fields of wildlife research and management, camera devices and photogrammetry have become an integral part of the toolkit for exploring otherwise‐unseen aspects of the biology, behavior, and control of the invasive brown treesnake (Boiga irregularis) on Guam. Because brown treesnakes are cryptic and nocturnal, and nearly all aspects of their ecology are influenced by snake size, methods are needed to estimate snake size from images captured by infrared wildlife cameras. Unfortunately, it is difficult to capture images of an entire snake’s length at a controlled distance from a simple camera setup. Here, I describe the allometric relationships between brown treesnake body length and potential predictors: head measurements, sex, and body condition. Head length (HL) was the most important predictor of body length, alone accounting for 95.9% of the variation in brown treesnake snout‐vent length (SVL). We provide simple regression equations for predicting brown treesnake length from head measurements, an example of how to extract measurements from images, and a convenient lookup table for predicting SVL and 80% prediction intervals from HL alone. Coupled with a simple camera setup that controls subject distance and includes size standards in the image, we can estimate brown treesnake body size from images that include only the head when photographed from above. These methods have been developed to enable ongoing assessments of brown treesnake predation risk following landscape‐scale suppression efforts that could enable the reintroduction of extirpated native wildlife.
... The types of measurements made required that the snake, and not a representation of the snake (i.e., a xerox copy, tracing, or a digital photo; e.g., Penning et al. 2013;Mangiacotti et al. 2014), be measured. The end of the retroarticular process, which was used to define both HL and LJL, cannot be seen and must be found by probing. ...
... In the case of the Cuban Boa mentioned above (Henderson and Arias 2001), the value of 4851 could easily have been 4800 or 4900. In the absence of repeated measurements, the range of possible sizes is impossible to guess, but the use of typological approaches, such as single digital images (e.g., Penning et al. 2013), will not solve this problem. A more fruitful approach to understanding snake sizes is to accept the range of sizes gained from repeated measurements and report the mean and standard deviation rather than a single figure or give the percentage of error in precision of the measurer, as determined by prior testing. ...
Article
Full-text available
We tested the precision and accuracy of common measurements of snakes by repeated measurement of the heads and trunks of 10 preserved snakes and 10 live snakes by two groups of five people over 10-wk periods. The measurements produced values with variances and ranges related to the nature of the variable, the measurer, and the snake, but accuracy could not be determined. Reporting sizes of snakes to high levels of accuracy is therefore unwarranted. Measurements of head variables on preserved and anesthetized live snakes had similar levels of variance that approximate half the variance of the same measures on live, unanesthetized snakes. Conversely, measurements of snout-vent length (SVL) on both preserved and unanesthetized live snakes had about twice the variance of the same measures made on anesthetized snakes. Measurers differed for all measurements of preserved snakes and for all head measurements of live, unanesthetized snakes, more experienced measurers generally yielding higher precision. Conversely, measurers did not differ for most measures of anesthetized snakes. Our data support suggestions that the most repeatable measures of SVL are made on anesthetized snakes. Lengths of the head and lower jaw can be measured with relative precision on a snake in any condition. Head width and supralabial length have both inter- and intrameasurer variances high enough to make them unreliable measures of head size. We conclude that features of live snakes most commonly measured vary because they have no exact size. We therefore suggest a new convention for reporting sizes of snakes.
... The types of measurements made required that the snake, and not a representation of the snake (i.e., a xerox copy, tracing, or a digital photo; e.g., Penning et al. 2013;Mangiacotti et al. 2014), be measured. The end of the retroarticular process, which was used to define both HL and LJL, cannot be seen and must be found by probing. ...
... In the case of the Cuban Boa mentioned above (Henderson and Arias 2001), the value of 4851 could easily have been 4800 or 4900. In the absence of repeated measurements, the range of possible sizes is impossible to guess, but the use of typological approaches, such as single digital images (e.g., Penning et al. 2013), will not solve this problem. A more fruitful approach to understanding snake sizes is to accept the range of sizes gained from repeated measurements and report the mean and standard deviation rather than a single figure or give the percentage of error in precision of the measurer, as determined by prior testing. ...
Article
We tested the precision and accuracy of common measurements of snakes by repeated measurement of the heads and trunks of 10 preserved snakes and 10 live snakes by two groups of five people over 10-wk periods. The measurements produced values with variances and ranges related to the nature of the variable, the measurer, and the snake, but accuracy could not be determined. Reporting sizes of snakes to high levels of accuracy is therefore unwarranted. Measurements of head variables on preserved and anesthetized live snakes had similar levels of variance that approximate half the variance of the same measures on live, unanesthetized snakes. Conversely, measurements of snout-vent length (SVL) on both preserved and unanesthetized live snakes had about twice the variance of the same measures made on anesthetized snakes. Measurers differed for all measurements of preserved snakes and for all head measurements of live, unanesthetized snakes, more experienced measurers generally yielding higher precision. Conversely, measurers did not differ for most measures of anesthetized snakes. Our data support suggestions that the most repeatable measures of SVL are made on anesthetized snakes. Lengths of the head and lower jaw can be measured with relative precision on a snake in any condition. Head width and supralabial length have both inter- and intrameasurer variances high enough to make them unreliable measures of head size. We conclude that features of live snakes most commonly measured vary because they have no exact size. We therefore suggest a new convention for reporting sizes of snakes.
Article
Full-text available
[The spring presence of two individuals of the Sea lamprey, Petromyzon marinus, in the River Mignone near Tarquinia (Northern Lazio) could highlight a new Italian reproductive site of this rare and endangered species. This exceptional possibility could certainly be favored by the good quality of both the waters of the Mignone, and the environmental context of the record, but would require the urgent equipment of the barrier of Le Mole with a fish ladder in order to allow the sea lamprey’s upstream migration towards the areas of the upper course, even more suitable for their reproduction]. [Article in Italian]
Article
To study sources of error in craniometric data and problems associated with using measurements taken by different observers, a sample of 10 bat skulls was measured with calipers three times by each of seven different individuals. Both intra- and inter-observer errors significantly contributed to the total variance of some characters. Intra-observer variability was responsible for ≤17% of the total variance (median = 10%) whereas inter-observer variability reached 30.7% (median = 3.9%). Characters with high intra-observer variability were those that exhibited high inter-observer variability. In comparisons involving measurements taken by different observers, a considerable risk exists that statistically significant results may arise due to measuring artifacts alone, especially when using large samples. However, measurements taken by different observers can be used in a single analysis if potentially risky characters are excluded from consideration. These include characters with high intra-observer variability, small means, or morphologically unclear end-points.