Less need for differentiation? Intestinal length of reptiles as compared to mammals.

Although relationships between intestinal morphology between trophic groups in reptiles are widely assumed and represent a cornerstone of ecomorphological narratives, few comparative approaches actually tested this hypothesis on a larger scale. We collected data on lengths of intestinal sections of 205 reptile species for which either body mass (BM), snout-vent-length (SVL) or carapax length (CL) was recorded, transforming SVL or CL into BM if the latter was not given, and analyzed scaling patterns with BM and SVL, accounting for phylogeny, comparing three trophic guilds (faunivores, omnivores, herbivores), and comparing with a mammal dataset. Length-BM relationships in reptiles were stronger for the small than the large intestine, suggesting that for the latter, additional factors might be relevant. Adding trophic level did not consistently improve model fit; only when controlling for phylogeny, models indicated a longer large intestine in herbivores, due to a corresponding pattern in lizards. Trophic level effects were highly susceptible to sample sizes, and not considered strong. Models that linked BM to intestine length had better support than models using SVL, due to the deviating body shape of snakes. At comparable BM, reptiles had shorter intestines than mammals. While the latter finding corresponds to findings of lower tissue masses for the digestive tract and other organs in reptiles as well as our understanding of differences in energetic requirements between the classes, they raise the hitherto unanswered question what it is that reptiles of similar BM have more than mammals. A lesser effect of trophic level on intestine lengths in reptiles compared to mammals may stem from lesser selective pressures on differentiation between trophic guilds, related to the generally lower food intake and different movement patterns of reptiles, which may not similarly escalate evolutionary arms races tuned to optimal agility as between mammalian predators and prey.

Introduction

There is a long tradition in ecomorphology of linking variation in digestive tract anatomy in vertebrates to the natural diets of the species, with the typical claim that herbivores have more complex, more voluminous, and also longer gut sections than carnivores [1, 2]. Given the widespread acceptance of this assumption, the paucity of actual statistical tests is surprising. In mammals, for example, statistical associations between intestinal length and diet had not been demonstrated until recently [3].

In groundbreaking work on 40 reptile species, Lönnberg [4] compared the snout-vent length to intestine length, showing that herbivores had an intestinal length of 293% relative to the snout-vent length, whereas omnivore and carnivore species had 184% and 131%, respectively. These findings shaped our understanding of reptile intestinal macroanatomy [5-7]; they have, however, not been assessed statistically.

Based on visual comparisons and analyses of a few species, reptiles have shorter digestive tracts than mammals [8-10]; they have significantly less digestive tissue mass [11]. This is generally explained by their lower metabolism and lower food throughput compared to endotherms [2, 8]. A lesser food throughput might, however, reduce the pressure on the reptilian digestive tract to be strictly associated with the peculiarities of a specific diet. Together with a lesser dependence on movement and agility compared to endotherms [12, 13], there might be less pressure to streamline aspects of the body bauplan. This rationale has been used to explain the seeming absence of a difference in torso volume in a small dataset between faunivorous and herbivorous reptiles, when such a difference is evident in mammals [14]. Visual impressions of digestive tracts of faunivorous vs. herbivorous species support the notion that differences between reptile trophic groups may be less distinct than they are in mammals (Fig 1). Therefore, we collated intestinal length data for reptiles of known body mass, to assess general scaling relationship, the comparison with mammals, and the influence of trophic level on these measures.

Fig 1

Examples of digestive tracts of faunivorous and herbivorous mammals as compared to reptiles.

Modified from Stevens and Hume [9]. Note the seemingly less distinct differentiation in intestinal length across trophic groups in reptiles.

Materials and methods

Relevant publications were collated by first using the sources of Franz et al. [11] as a starting point, but also by performing a literature research in the three search engines Google scholar, PubMed, and Web of Science. Search terms included ‘intestine’/’gastrointestinal tract’, ‘length’/’morphology’, ‘reptile’/scientific names of reptile genera. Reference lists, and ‘cited by’ lists, of publications thus identified were also used. Sources were only included if they provided the species, body mass (BM) or snout-vent-length (SVL) or carapace length (CL) of the individuals studied, as well as measurements of all or some of the gut sections (total intestine–TI, small intestine–SI, large intestine–LI including the colon and the caecum). Additional unpublished data was obtained from post-mortem examinations carried out by MSE and MC over the last decades. Because in many sources, the caecum was not indicated separately but as part of the LI measurement, the colon was not recorded separately. When the body mass of the individuals was not available from the source but only SVL or CL was given, body mass was estimated from SVL or CL using published allometric equations [15-17]. For Podocnemis spp., no corresponding equation was published; therefore, we derived our own BM-CL regression equation based on data from various sources [18-24].

Weighted species means (correcting for sample size) were calculated of each morphometrical parameter and the corresponding body mass. For example, if more data was available for small intestine than for caecum length of a species, then the body mass used for associations with small intestine length was different from the one used in the same species for associations with caecum length. Data on the natural diet of reptiles was collated from the literature; species were classified as faunivores, omnivores or herbivores. To our knowledge, no continuous measure of reptile diets is available. The referenced dataset is available as an online supplement. Mammal data used for comparison was taken from Duque-Correa et al. [3].

The phylogenetic tree for reptiles was built from the supertrees (based on molecular data and time calibrated by the fossil record) for Squamates [downloaded from http://vertlife.org/phylosubsets/, following 25] and for Testudines [26]. Trees were pruned in R using scripts from the library ‘ape’ [27] and ‘tidyverse’ [28] in order to obtain a final tree inclusive only of the species for which GIT data and body masses were available. For mammals, the phylogenetic tree from Duque-Correa et al. [3] was used; the reptile and mammal trees were merged manually following the amniote topology from Fig 1 in Gemmell et al. [29]. The two crocodilian species were added manually as a sister group of the testudines. Divergence times between all major manually merged clades were set using TimeTree.org, which provides an estimated time of divergence. This was also used to separate Crocodylus from Alligator (80.0 Ma). Thus, the split between mammals and sauropsids (root of the tree) was set at 318 Ma. The separation of lepidosaurids (Squamates) from the clade of Testudines and Crocodylia was set at 280 Ms while the internal split between Testudines and Crocodylia was set at 254 Ma. The emergence of Squamates was set to 180.57 Ma.

Statistical analyses were done on (i) all available data (i.e., at different sample size for the different intestine sections–generally larger samples for the total intestine than for individual sections), and on a subset that comprised (ii) those species for which both small and large intestine length was available. Additionally, analyses were performed on the taxonomic levels of squamates, lizards, snakes, and turtles. First, the allometric relationships with body mass were determined, and it was assessed which intestine section showed the best fit with body mass. Scaling exponents were termed ‘more’ or ‘less than geometric’ if they were above or below the expected isometry of 0.33 [30]. Then, the effect of diet was evaluated.

Allometric regressions were performed as linear regressions on log-transformed data, because we are not aware of another method to which we can apply phylogenetic generalized least squares. All analyses were performed using generalized least squares (GLS) phylogenetic generalized least squares (PGLS), recording the 95% confidence interval for parameter estimates, using the R packages ‘caper’ [31] and ‘nlme’ [32]. The phylogenetic signal lambda (λ) was estimated by maximum likelihood. The significance level was set to 0.05. Different models applied to a certain dataset (separately for GLS and PGLS) were compared using the small sample corrected Akaike’s information criterion (AICc) [33], considering models that differed by more than 2 (ΔAICc > 2) as providing a different fit to the data.

Results

Intestinal length information was available for the total intestinal tract (157 species), the small intestine (147 species), the large intestine (141 species), and the caecum (52 species). Generally, the small intestine represented the longest intestinal section, followed by the large intestine and the caecum (Fig 2). In analyses using phylogenetical generalized least squares (PGLS), the phylogenetic signal λ varied across its whole range from 0 to 1, depending on the specific dataset; generally, larger datasets had a higher λ (Table 1, S1-S8 Tables in S1 File).

Fig 2

Magnitude comparison of the length of the small intestine, the large intestine and the caecum in reptiles.

Note that the slope of the scaling of the three sections in log-log-space appears similar.

Table 1

Summary statistics for allometric scaling of insteinal length with body mass in reptiles, as log(y) = a + b log(body mass), or y = (10a) BMb (significant parameters in bold).

				GLS		parameter	PGLS			parameter
Dependent	Model	n		AICc	ΔAICc	(95%CI)	lambda (95%CI)	AICc	ΔAICc	(95%CI)
All data
Total intest.	BM	157	a	-	-	1.82 (1.79 to 1.86)	0.77 (0.56 to 0.90)	-	-	1.77 (1.61 to 1.92)
			b			0.40 (0.38 to 0.42)				0.38 (0.35 to 0.4)
Small intest.	BM	147	a	-	-	1.66 (1.62 to 1.70)	0.50 (0.22 to 0.74)	-	-	1.62 (1.49 to 1.75)
			b			0.38 (0.36 to 0.41)				0.37 (0.33 to 0.40)
Large intest.	BM	141	a	-	-	1.24 (1.19 to 1.29)	0.56 (0.31 to 0.75)	-	-	1.14 (0.98 to 1.31)
			b			0.43 (0.40 to 0.46)				0.40 (0.37 to 0.44)
Caecum	BM	52	a	-	-	0.48 (0.37 to 0.59)	0.00 (NA to 0.27)	-	-	0.48 (0.37 to 0.59)
			b			0.42 (0.35 to 0.49)				0.42 (0.35 to 0.49)
Consistent data (species for which both small and large intestinal length are available)
Small intest.	BM	141	a	-82.3	0.0	1.64 (1.60 to 1.68)	0.56 (0.31 to 0.75)	-125.5	0.0	1.61 (1.46 to 1.75)
			b			0.38 (0.35 to 0.40)				0.36 (0.33 to 0.39)
Large intest.	BM	141	a	-26.5	55.8	1.24 (1.19 to 1.29)	0.57 (0.33 to 0.76)	-61.7	63.8	1.14 (0.98 to 1.31)
			b			0.43 (0.40 to 0.46)				0.40 (0.37 to 0.44)

NA no model output.

Allometry

In most datasets analyzed, intestinal lengths scaled more-than-geometrically (positive allometry) at an exponent whose 95% confidence interval was above 0.33. Regardless of the phylogenetic signal, the simple scaling relationships were generally similar in generalized least squares (GLS) and PGLS (Table 1). When using only species for which all respective data were available, the small intestine-body mass relationship showed a better fit that the large intestine-body mass relationships (ΔAICc GLS = 56, PGLS = 63), suggesting that the large intestine is more subjected to additional influence factors (Table 1). Body mass was part of all subsequent models.

Taxonomic comparisons

Based on body mass, no differences in the scaling between the reptile groups (lizards, snakes, turtles) was evident (S1 Fig and S1 Table in S1 File), even though numerically, snakes had the lower and turtles the higher scaling exponents.

Compared to mammals, reptiles generally had shorter intestines at comparable body mass (Fig 3, S2 Table in S1 File). The models including only body mass but not the taxonomic group (mammal / reptile) always had the lowest support (ΔAICc >150). Using the difference in the GLS models, reptile intestines are only 30–40% of the length of mammalian intestines at similar body mass. In GLS, both the taxon group (mammal / reptile) and the body mass-taxon group interaction were generally significant, except for the large intestine, where the interaction was not. Consequently, the model including the interaction had the best support (ΔAICc to other models: total intestine 9.7, small intestine 10.1, caecum 17.0), except for the large intestine where the model that only included the taxon group was best supported (ΔAICc to other models 7.0). Hence, while the scaling for the large intestine was similar between mammals and reptiles (exponent 95% confidence intervals 0.39–0.44 vs. 0.40–0.46, respectively), the scaling of the total and small intestine was steeper in mammals than in reptiles (at exponents for the total intestine of 0.46–0.49 vs. 0.38–0.42, small intestine of 0.44–0.47 vs. 0.36–0.41, respectively). For the caecum, scaling in mammals was lower than in reptiles (0.18–0.24 vs. 0.35–0.49, respectively).

Fig 3

Relationship of body mass, intestinal length and trophic level in reptiles for (A) total intestine (n = 157 species), (B) small intestine (n = 147), (C) large intestine (Caecum and colon) (n = 141), (D) caecum (n = 52), as compared to mammals (light grey symbols). For statistics, see S3-S5 Tables in S1 File. Mammal data from [3].

In PGLS, there was a strong phylogenetic signal when analyzing mammals and reptiles together (λ ≧ 0.90). In PGLS, taxon group (mammal / reptile) was never significant, and the body mass x taxon group interaction was only significant for the caecum. Models that included taxon group (with or without interaction) were never better supported than the models with body mass only (ΔAICc always ≦ 2; S2 Table in S1 File).

Trophic level

Adding the categorical information on trophic level (fauni-, omni-, herbivore) to the models for all reptiles did not result in an improved data fit in GLS, and in PGLS only in the case of the large intestine (with ΔAICc to the model without trophic group = 2.4; S3 Table in S1 File); visually, scatter plots did not indicate a systematic effect of trophic level (Fig 3). In GLS, trophic level was significant for the large intestine, with herbivores having a 1.3 times longer large intestine than faunivores. This difference was also supported in PGLS (S3 Table in S1 File). When making the same assessment for only those reptile species for which both small and large intestine length was known, the result was the same. Again, the small intestine-body mass relationship had a better fit than the large intestine-body mass relationship, and the addition of trophic level did not improve model fit in GLS, and in PGLS only barely for the large intestine (ΔAICc to the model without trophic group = 2.4 (S3 Table in S1 File)).

When considering reptile taxa individually, adding trophic level did not improve model fits in Squamates or Testudines (only for the Squamates large intestine in PGLS, S4 Table in S1 File). However, in the lizards, it improved model fit for the total intestine (GLS ΔAICc = 2.9, PGLS ΔAICc = 8.3), and in PGLS also for the small and the large intestine (ΔAICc 2.8 and 3.8, respectively). In both models, herbivorous lizards had significantly longer measures than faunivorous ones (S4 Table in S1 File). Because snakes only included faunivores, they were not assessed separately for trophic level.

In the large-scale comparisons with mammals, GLS models that included both taxonomic group (mammal / reptile) and trophic level were the best supported for total intestine, large intestine and caecum (ΔAICc > 84), and among the best supported for the small intestine (S5 Table in S1 File). Reptiles had significantly shorter, and herbivores longer intestines in all these models. In PGLS, taxonomic group was never significant and did not improve model fit; trophic level was included in the best models for total intestine, large intestine and caecum, and herbivores always had significantly longer measures. By contrast, the best models for the small intestine did not include trophic level in PGLS (S5 Table in S1 File).

Body mass vs. body length scaling

Given that the original concept of intestinal length being linked to diet in reptiles originated from Lönnberg [4], who did not use body mass but snout-vent-length (SVL) as the basis for the comparison, we re-analyzed this individual dataset (n = 40 species, total intestine length) with and without the trophic level information provided by that author. In both GLS (ΔAICc = 9.7) and PGLS (ΔAICc = 3.2), trophic level addition improved model fit, and while herbivores only had a significantly longer total intestine in GLS but not PGLS (95% CI of parameter estimate 0.24 to 0.58 and -0.02 to 0.46, respectively) than faunivores, omnivores had significantly longer measures than faunivores in both models (0.05 to 0.33 and 0.04 to 0.29, respectively). In these models, total intestine length scaled to SVL at an exponent that included linearity (GLS: 0.64 to 1.11, PGLS: 0.87 to 1.35).

Based on these observations, comparisons were repeated for the dataset that was based originally on SVL. In all reptiles, adding trophic level to the SVL-based GLS and PGLS models increased model fit for the total, small and large intestine, and for the caecum in PGLS (ΔAICc to the model without trophic level GLS > 24, PGLS > 3; S6 Table in S1 File). Models always indicated longer measurements in herbivores compared to faunivores. Models that did not include trophic level but whether a species was a snake or not had even better model fits in GLS for the total and small intestine and the caecum (ΔAICc to trophic level models GLS > 6) but not for the large intestine (S6 Table in S1 File), indicating that snakes are not only peculiar in being exclusively faunivores, but also either with respect to intestine length or to SVL. A dataset comprising only lizards and turtles led to identical results (except that the factor ‘snake’ was not assessed; S7 Table in S1 File). In the lizard only dataset, trophic level always not improved model fit (ΔAICc GLS > 7, PGLS > 2.6; S7 Table in S1 File). For turtles, trophic level improved model fit only in GLS for the large intestine (ΔAICc = 11.9; S7 Table in S1 File), with herbivores having longer measures, but not in PGLS.

When comparing model fit in the dataset for the total intestine length in which both body mass and SVL could be used as a basis for scaling, the body mass-based model had the better fit (ΔAICc GLS 63, PGLS 32), and this difference was maintained when other factors were added. This is due to the difference in the scaling in snakes, which have a similar intestine length as other reptiles for their body mass (S1 Fig in S1 File), but–predictably–a shorter intestine length than other reptiles for their body length (Fig 4). Adding being a snake or trophic level information to GLS models improved model fit similarly, both for body mass and SVL models (S8 Table in S1 File). In PGLS, being a snake did not add substantially to the model fit with body mass and trophic level (ΔAICc ≦ 2); for PGLS SVL models, trophic level and being a snake both contributed substantially to model fit. In the same dataset, SVL as dependent variable scaled with body mass in a geometric fashion (with an exponent whose 95% confidence interval included 0.33); the scaling model fit was dramatically improved when being a snake was included (ΔAICc GLS 109, PGLS 38), so that at the same body mass, snakes had a 4–4.5 times higher SVL than other reptiles. Trophic level did not improve this scaling (S8 Table in S1 File).

Fig 4

Relationship of snout-vent-length (SVL) and intestinal length in reptiles for (A) the total intestine, (B) the small intestine, (C) the large intestine, (D) caecum, for taxonomic groups. Note that snakes are all faunivorous and have, at a similar intestine length, a similar body mass as other reptiles, but a longer body length. For statistics, see S8 Table in S1 File. The dotted line represents y = x to facilitate gauging the slope of the length-length relationships.

Discussion

The present study provides a comprehensive data collection on reptilian intestinal length, which indicates that a distinct association with trophic niche is not as evident in reptiles as in mammals. The results show that it is important to consider different sections of the intestinal tract separately when possible, most likely because they fulfil different functions. Even though our data collection represents possibly the largest one available to date, presented also in the hope that it will be useful for other researchers, the limitations of both the dataset itself and the focus on a single kind of measurement–length measures–must be stated.

Constraints

Compared with mammals [3], distinctively fewer data on intestine length exist in reptiles that can be linked, at least, to snout-vent- (SVL) or carapax length (CL). Large collections on reptiles kept as pets either did not resolve their sample to the species level [34], or did not record a proxy (body mass, CL for turtles) that would make the data comparable [35]. Meiri [16] already argued that reptile studies should more often report the body mass of the investigated specimens, a claim which we support. Generating more data on reptile gastrointestinal morphology should be a feasible task, considering the occurrence of reptiles in zoological institutions or as pets.

Gastrointestinal dimensions and tissue mass can fluctuate distinctively in reptiles, especially in faunivorous species with an intermittent feeding regime [36, 37], or in animals undergoing seasonal dormancy [38]–factors that could not be accounted for in the present data collection. As in similar large-scale comparisons in mammals [3, 39], variation due to origin (free-ranging or captive), inter-individual differences due to actual diets consumed, tissue preservation or body condition could also not be accounted for. For example, already the original study by Lönnberg [4] does not indicate the origin, the actual diet, feeding or dormancy state of the specimens, nor the means of their preservation.

As already mentioned by Lönnberg [4], SVL may not represent an ideal proxy for body size because of the difference in body shape (slender vs. stout) between species. Transforming SVL or CL into body mass could dampen or exaggerate potential differences between trophic levels, if trophic level had an effect on the body mass-SVL relationships. This was not the case in the present dataset (S8 Table in S1 File). However, Meiri [35] reported the nearly significant effect (at P = 0.058) that herbivorous lizards were heavier for their SVL, or had a shorter SVL for their body mass–in other words, they had a stouter body shape. When estimating body mass from SVL using a general equation that does not account for trophic level, this would mean that herbivorous animals would be assigned a slightly underestimated body mass, which should make any putatively longer intestinal tract even more evident in comparative approaches. The fact that this was not the case suggests that the result of little large-scale trophic differentiation in intestinal length of the present study is robust in spite of the SVL-body mass transformation.

As described for different mammalian datasets [reviewed in 3], all intestinal lengths investigated in the present study scaled at a higher exponent than expected based on simple geometry (i.e., positive allometry at an exponent > 0.33). The only consistent exception to this were the snakes, whose intestinal scaling exponent always included 0.33 in the confidence interval (S1 Table in S1 File).

The current explanation for the positive allometry is that whereas gut surface scales geometrically with body mass, diffusion distance (i.e., gut diameter) should remain short, i.e. scale less-than-geometrically (with negative allometry), and hence gut length must necessarily scale more-than-geometrically to compensate [40]. This hypothesis would have to be tested in two ways: (i) with a large dataset on intestinal diameter; (ii) with a large dataset on intestinal surfaces, which would test the underlying assumption that this surface scales geometrically with body mass. Neither dataset exists or can be collated easily, to our knowledge, and thus resolution of these question hinges on the future measurement of anatomical data.

Based on the concept of geometric gut surface scaling, a possible reason for the lower scaling in snakes could lie in the very intermittent feeding pattern in the snakes [41, 42] with a particular dependence on gastric dilatation, but little dependence on a long intestine with a high throughput capacity.

Comparison with mammals

The present study impressively corroborated previous observations based on smaller sample sizes [8-10] that reptiles have shorter digestive tracts than mammals. This is explained by the lower metabolism of the poikilothermic reptiles as compared to homeothermic mammals [43, 44]. At least for herbivorous reptiles, both the wet gut contents and the calculated dry gut contents are of a similar magnitude as in mammals [11, 45], suggesting a similar capacity of the main digesta-holding gut compartments. Together with longer intestines in mammals, this would necessarily imply intestines of larger diameter in reptiles. Again, to our knowledge, a comparative dataset on intestinal diameter does not exist. On the other hand, in a comparison of two reptile and two mammal species, Karasov et al. [8] reported a lesser nominal (i.e., without villi) surface of the small intestine in the reptiles, suggesting that mammal intestines are not only longer to compensate for putatively smaller diameter, but actually have a larger surface and hence larger volume capacity. Without further measurements, this discrepancy remains unresolved.

On the basis of comparable body mass, reptiles have lower food intakes than mammals [8, 45]. Assuming a similar gut volume, this necessarily translates into a slower passage of digesta through the digestive tract [8, 45]. Mammals, with their higher energetic requirements, need to eat more, and consequently have a faster digesta passage. To avoid negative effects of this faster passage on digestive efficiency, mammals employ several strategies in parallel: they achieve a distinct particle size reduction during ingestion via chewing [46], as smaller particles can be digested at a much faster rate than larger ones [reviewed in 47]. They possibly have shorter diffusion distances in the intestines via smaller intestine diameters (as deducted from the similar gut capacity and the longer intestines demonstrated in the present study). And they possibly have more absorptive surface due to microanatomical structures, in particular in the small intestine; their intestine also achieves higher absorption rates [48], and their intestinal functions are evidently less dependent on ambient temperatures. Because of these (and probably other) adaptations, mammals achieve a similar and even higher digestive efficiency than reptiles in spite of their higher intakes [8, 45, 49, 50].

Whereas these considerations refer to the difference in the scaling factor (the ‘intercept’ in log-linear regressions), we also found a lesser scaling exponent (the ‘slope’ in log-linear regressions) for the total and the small intestine in reptiles as compared to mammals [3] (e.g., small intestine: GLS exponent 0.36–0.41 in reptiles vs. 0.44–0.47 in mammals). In PGLS, this difference was not significant (as indicated by the non-significant taxon-body mass interactions in S2 Table in S1 File) (small intestine: PGLS exponent 0.33–0.40 in reptiles as compared to 0.38–0.43 in mammals). This indicates that the seeming difference in slope is caused by the distribution of specific taxa across the dataset but not by a systematic difference between reptiles and mammals. By contrast, the difference in the scaling exponent for the caecum was also significant in PGLS (GLS exponent in reptiles 0.35–0.49 vs. 0.18–0.24 in mammals; PGLS in reptiles 0.35–0.49 vs. 0.25–0.32 in mammals). As previously, we suggest that this is due to particularly long caeca in small mammals, possibly associated with the widespread strategy of ‘colonic separation mechanism’-linked coprophagy in this group [3], and not due to a particularly steep scaling in reptiles (in which such a mechanism and the associated coprophagy have not been reported).

Individual studies comparing organ masses between selected mammal and reptile species of similar body mass indicate larger organ masses in mammals [e.g. 51, 52]. Larger literature compilations of data from several studies show that, for a similar body mass, reptiles have less gastrointestinal tissue [11; this study]. Additionally, while liver mass is probably similar in mammals and reptiles, heart and kidney mass might be slightly lower in reptiles [11]. This raises the interesting question what it is that reptiles have ‘more’ than mammals to achieve the ‘similar body mass’. To our knowledge, this question has not been answered so far, and we can only speculate that reptiles may have distinctively more body tissue in the form of their tails.

The general argument is that a less digestible diet, typically one of a higher proportion of plant material, requires a longer digestive tract [1, 53]. More specifically, in mammals it was the large intestine, not the small intestine, that showed this adaptation to a plant diet [3]. In his original study on lizards, Lönnberg [4] already described that it is particularly the large intestine that was longer in herbivorous species. By contrast, Herrel et al. [54] found that it was particularly the small intestine that was longer in four omnivorous vs. eleven faunivorous lizards, and hypothesized that faunivory (as opposed to any proportion of plant matter in the diet) should be particularly distinguishable by intestine length in lizards.

For reptiles, a large number of studies described particular adaptations to herbivory. These studies typically focus on the caeco-colonic dilatation and valve-like structures in lizards [4, 6, 54–56], but intestinal length has also been used [7, 57–62]. However, our data collection does not indicate clear differences in intestinal tract length between the trophic groups. The only indication was that the large intestine was significantly longer in herbivores by a factor of 1.3 (S3 Table in S1 File), compared to a GLS-derived factor of 1.9 in mammals [3]. The difference between GLS and PGLS in the reptiles suggests that whereas the pattern is not distinct when just comparing the trophic groups (Fig 2), the difference between the trophic groups is systematic within various taxa. In this dataset, these taxa belonged to the lizards. When lizards were analyzed alone, the longer large intestines (but not small intestines) in herbivorous species were detectable in both GLS and PGLS (S5 Table in S1 File).

In the dataset based on SVL measurements only, trophic level had an effect on intestinal lengths in the complete dataset, even when the taxonomic group of snakes was accounted for in the models, mainly due to a large intestine that was 1.4 times longer in herbivores than in faunivores (S6 and S7 Tables in S1 File). In the SVL dataset, the number of turtle species that could be included was distinctively higher than in the BM dataset (63 vs. 18 species). This was due to one investigation [35] that did not record BM or carapax length in turtles, but only SVL, making a transformation to BM impossible (because no transformation equations for SVL to BM exist for turtles). The fact that larger datasets lead to stronger trophic signals was already observed in mammals [3]. Therefore, data for more reptile species might well lead to a stronger signal on trophic specialization of intestinal length. Based on the available data, however, we conclude that a trophic signal in reptile intestinal length is weak.

We can only speculate about reasons for a putatively less distinct effect of trophic level on intestine length in reptiles than in mammals. Although to our knowledge, a comparative dataset is missing that assesses the relative food intake, digesta passage and digestibility in herbivorous vs. faunivorous reptiles, we assume that, as in mammals [63], herbivorous reptiles should generally have a higher food intake than faunivorous ones. However, with reptile energy expenditures [44] and food intake levels [45] of approximately only 10% of that of mammals, the basic reptilian intestinal tract might not be as challenged as that of mammals with respect to optimization of digestion and agility. A lack of requirements for agile locomotion might contribute to the difference in mammals. Possibly, a lower proportion of faunivorous reptiles are active pursuit hunters, with ambush predation considered the ancestral state [64]. Insectivorous forms employ a sit-and-wait or a search foraging strategy [65, 66], and snakes that include relatively large vertebrate prey in their spectrum are typically sit-and-wait predators [67]. In mammals, the clear difference in intestinal tract length [3] as well as in abdominal cavity volume [14] between herbivores and faunivores may well derive from an arms race in agility [13, 68–70], where in particular the faunivores reduced intestinal tissue and body shape to maximize pursuit efficiency. Possibly, reptiles additionally move less within a comparable time frame than mammals [12] in spite of similar home range sizes [71]. Comparative evaluations of reptile and mammal movement and hunting patterns, as well as comparative datasets on digestive performance of reptilian herbivores and faunivores, are required to further test these hypotheses. The less differentiated reptile gastrointestinal anatomy might also correspond to the finding that the gastrointestinal microbiome shows a less stringent phylogenetic signal in reptiles than in mammals [72].

Body shape

The finding that intestine length is similar across reptile groups when compared to body mass but not to snout-vent-length is due to the special position of the snakes (Fig 4). Clearly, the elongation in body length in snakes is not accompanied by a corresponding elongation of the intestine, even though other organs, such as the stomach, the kidneys, lungs or spleen clearly appear elongated when compared to other reptiles [73]. This contrasts with the mustelids–a group of mammalian carnivores that also evolved an elongated body shape [74]. At comparable body mass, mustelids have longer intestines than other carnivores [75], mirroring a comparison of body mass and body length [76]. Thus, the hypothesis results that different mechanisms of body elongation occur in snakes and mustelids–resulting in different effects on the intestinal tract. More studies, including embryological ones, are required to address this difference.

Conclusions

The results of the present study indicate that while a link between trophic niche and intestinal length can be detected in some datasets, and in particular for the large intestine, this link appears weaker in reptiles compared to mammals, which might be linked to a lesser need for differentiation in reptiles because of their lower metabolism, lower food intake, and less agile movement patterns. Clearly, more data are required to fully address these questions. In particular, future studies could fruitfully combine a variety of measurements, not only intestinal lengths, but organ tissue and contents mass, diameters and surfaces, as well as both body mass and snout-vent-length or carapax length, to facilitate more comprehensive comparisons that have to rely less on inductive speculation.

Supplemental information.

S1-S8 Tables & S1 Fig.

(DOCX)

Click here for additional data file.

Phylogenetic tree.

Reptiles & mammals.

(TXT)

Click here for additional data file.

Reptile intestine data.

Full dataset.

(XLSX)

Click here for additional data file.

Lönnberg dataset.

(XLSX)

Click here for additional data file.

Podocnemis equation.

Data for regression line.

(XLSX)

Click here for additional data file.

Data sources.

Reference list.

(DOCX)

Click here for additional data file.

18 May 2021

PONE-D-21-11487

Less need for differentiation? Intestinal morphology of reptiles as compared to mammals

PLOS ONE

Dear Dr. Clauss,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

I am happy if general morphological-physiological trends in animals are being analyzed, However, there are few strong conclusions derived from your data. Please look at the comments below. Hope you can improve the manuscript.

Please submit your revised manuscript by Jul 02 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ulrich Joger

Academic Editor

PLOS ONE

Additional Editor Comments:

The reviewer I found for your manuscript found nothing “wrong” in this paper with methods or statistics, and the writing appeared clear. Yet he proposed to reject it. I am not so hard and like to give you a Chance to improve the paper. Please enlarge the data (if you have more) or Change your conclusions by carefully differentiating between hard evidence and speculation. It is not a shame if a hypothesis is not supported by a data set, but a discussion must be self-critical.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating in your Funding Statement:

[This study is part of Swiss National Science Foundation (http://www.snf.ch) project CRSII5_189970 / 1 (to MC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.].

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

3. Please amend either the title on the online submission form (via Edit Submission) or the title in the manuscript so that they are identical.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript analyzes and compares in reptiles and terrestrial mammals variation in intestine length in relation to body size, diet, and phylogeny. The statistical analysis of the data seem fine, and the writing is clear.

My main critique is that the paper advances knowledge about digestion or ecomorphology very little because it focuses on one thing, really one dimension, intestine length. As a reader, I approach it with about as much interest as I would a thorough analysis of some other intestinal analogue such as intestine width. Width could be analyzed in the same way as length is here, but who would be interested in that? Put the two together and one would certainly have a lot more, such as surface area that can be related to substrate absorption or niche space for adherent microbes, or gut volume that can be related to holding capacity for digesta or microbes. Add time (another dimension) and one can make inferences about turnover kinetics and the matching of rates such as digesta breakdown or metabolic demand. But working with gut length (or width) alone limits saying much definitive about function. The authors recognize this (i) by pointing out that most authors who consider herbivorous reptile gut ecomorphology focus on the caeco-colonic dilatation and valve-like structures (lines 369-371), and (ii) with statements that additional data on gut diameter, which are necessary to make stronger inferences related to gut length, are not available (line 299,314-320, 329). Consequently, there are few conclusions I would call reliable, some seem poorly supported by data, and many points of discussion seem thin and speculative.

For example, this seems to be the case for every point discussed in the section “Comparison with mammals” (lines 306-358). Because the issue of differences in gut diameter is unresolved, the authors try to infer it using an inconclusive argument (lines 310-320) followed finally by invoking the idea that mammals do have smaller intestine diameters (lines 329-330), which they do not have strong evidence for.

As another example, they speculate about differences in scaling exponent of intestine length (beginning line 38), which is perhaps not even meaningful without knowing intestine diameter and hence actual surface area. According to Karasov and Hume (1997) there is no difference in the scaling exponent of intestine nominal surface area between reptiles (5 species) and mammals (20 species). That’s not a lot of evidence either, so if the science needs more data then the best solution is to provide more data.

At lines 353-358 – I do not see why the authors state that liver size is probably similar in mammals and reptiles. The data from Else and Hulbert (1985) suggest otherwise, as do the regression lines shown in Fig. 1c of Franz et al. (2009). The authors finish this section with the sentence “we can only speculate that reptiles may have distinctively more body tissue in the form of their tails”.

Testing putative trophic differences in intestine length was the “raison d’etre” for the paper (lines 26-28; 51-55). In the Discussion of this (lines 360-365) they point out that in mammals it applies mainly to the large intestine, which they also cite as Lonnberg’s conclusion for reptiles >100 years ago. It is what the authors found again here, mainly in lizards. The advance in knowledge that they offer is that the trophic signal might be weaker in reptiles than in mammals, though “larger datasets lead to stronger trophic signals” (lines 390-391). So, they end this section of the Discussion with speculation “…about reasons for a putatively less distinct effect of trophic level on intestine length in reptiles than in mammals” (395-396).

I reiterate that I found nothing “wrong” in this paper with methods or statistics, and the writing is clear. I mainly fault it for its narrow focus that to my thinking leads to conclusions poorly supported by data, and little advance in more reliable understanding, as opposed to speculation.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

21 May 2021

please see the detailed reply letter

Submitted filename: ReplyLetter_210519_mc.docx

Click here for additional data file.

31 May 2021

Less need for differentiation? Intestinal morphology of reptiles as compared to mammals

PONE-D-21-11487R1

Dear Dr. Clauss,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ulrich Joger

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Your answers to the critique by the reviewer convinced me that there is not really much you can do to improve your manuscript - only add some words in the discussion explaining what the limits of your data are and what should be done in the future. As you have done this to my satisfaction, I accept your manuscript now. Just put genus names like Crocodylus and Alligator in italics.

Reviewers' comments:

25 Jun 2021

PONE-D-21-11487R1

Less need for differentiation? Intestinal length of reptiles as compared to mammals

Dear Dr. Clauss:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Ulrich Joger

Academic Editor

PLOS ONE