Balancing Acts: Multi-level physical interactions provide insights into snake climbing behavior. (Scaling the scales of scaling scales) Open Access
Riiska, Calvin (Summer 2026)
Abstract
Environmental forces are a crucial piece to understanding animal locomotion. As animals move through their environment, they must contend with several complex substrates while their body acts as an interface with which to interact with them. Here we particularly focus on the physical interactions involved in climbing behavior in snakes. Snakes experience unique challenges when locomoting because they are limbless. However, they have adapted body manipulation strategies and physical characteristics that help them navigate these interactions. Arboreal environments are particularly extreme because animals must manage tree structures of varying diameter, roughness, and flexibility all while contending with gravity. Failure to locomote successfully can be catastrophic due to falling from great heights.
A short project on katydid locomotion probed gait changes that occur when walking up a steep incline. Katydids take on gaits with 4-5 legs in contact with the substrate as the incline angle increases to gain added stability as opposed to their normal tripod gait at low inclines.
Surprisingly, snakes can climb flat and low curvature surfaces where strategies like surface constriction are impossible. We investigated how cornsnakes manipulate surface features on such a surface to promote success ascending and descending. We find that computationally, purely dissipative interactions at contact points allows for successful climbing, as does simple wave propagation on a robotic model. However, when the cornsnakes climb, descents proceed as predicted while in ascents, the snakes actively redirect force at contacts. They use redundancy provided by many contacts to rearrange force in a network and access multiple solutions to satisfy force balance.
Whole body movements are often used to describe an animal’s locomotion, but microscopic structures on an animals’ surface may help tune interactions. Here we investigate ventral microstructures on vipers and their relation to different locomotor behaviors. We find that preserved museum specimens retain microstructural features similar to shed skins which increases the availability of samples to study. We leverage this result to find a third instance of a convergently evolved microstructure in sidewinders. We close by presenting preliminary work on what microstructural features may be common and beneficial to arboreal snakes versus terrestrial ones.
Table of Contents
List of Figures v
List of Tables xxvii
1 Introduction 1
1.1 Snake locomotion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Climbing in limbed and limbless animals . . . . . . . . . . . . 4
1.2 Microstructures on snakes . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Dissertation structure . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 Physical principles applied to animal behavior 12
2.1 Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.1.1 Movement in fluids . . . . . . . . . . . . . . . . . . . . . . . . 15
2.1.2 Movement in sand . . . . . . . . . . . . . . . . . . . . . . . . 23
2.1.3 Movement on solid substrates . . . . . . . . . . . . . . . . . . 26
2.2 Functional textures . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.2.1 Mechanical interactions between surfaces . . . . . . . . . . . . 30
2.2.2 Interactions with water . . . . . . . . . . . . . . . . . . . . . . 32
2.2.3 Interactions with light . . . . . . . . . . . . . . . . . . . . . . 33
2.3 Discussion and Open Questions . . . . . . . . . . . . . . . . . . . . . 35
2.4 Applications to the research at hand . . . . . . . . . . . . . . . . . . 39
3 Arboreal katydids exemplify beneficial gait changes 41
3.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.3.1 Specimen Collection and Experimental Protocol . . . . . . . . 45
3.3.2 Tracking and Gait Analysis . . . . . . . . . . . . . . . . . . . 47
3.3.3 Statistical Analysis . . . . . . . . . . . . . . . . . . . . . . . . 49
3.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.S Supplementary Material . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.S.1 Species Identifications: . . . . . . . . . . . . . . . . . . . . . . 63
4 Mechanics of snake climbing behavior 65
4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.3 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.3.1 Study subjects . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.3.2 Experimental apparatus . . . . . . . . . . . . . . . . . . . . . 69
4.3.3 Experimental procedure . . . . . . . . . . . . . . . . . . . . . 70
4.3.4 Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.4.1 Climbing is quasi-static. . . . . . . . . . . . . . . . . . . . . . 75
4.4.2 Computational and robotic models establish a minimal climbing
baseline. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.4.3 Snakes actively redirect forces during ascents. . . . . . . . . . 82
4.4.4 New contacts trigger stereotyped, balance-preserving force re-
distribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4.4.5 Body segments are recruited for rearrangement based on the
environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
4.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.S Supplementary Material . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.S.1 Acrylic surface with no posts . . . . . . . . . . . . . . . . . . . 92
4.S.2 Trial counts by individual . . . . . . . . . . . . . . . . . . . . 94
4.S.3 Bracing force . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.S.4 Forces in the third dimension . . . . . . . . . . . . . . . . . . 95
4.S.5 Metrics relative to contact location . . . . . . . . . . . . . . . 96
4.S.6 Nullspace calculation . . . . . . . . . . . . . . . . . . . . . . . 99
4.S.7 Bootstrapping fit parameters . . . . . . . . . . . . . . . . . . . 100
4.S.8 Measuring snake length . . . . . . . . . . . . . . . . . . . . . . 101
4.S.9 Force sensor calibration . . . . . . . . . . . . . . . . . . . . . 101
4.S.10 Randomizing trials . . . . . . . . . . . . . . . . . . . . . . . . 104
4.S.11 Contact triggered average plots . . . . . . . . . . . . . . . . . 104
5 Snake Microstructures – convergently evolved morphology 106
5.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.3.1 Sample preservation . . . . . . . . . . . . . . . . . . . . . . . 112
5.3.2 Sample preparation . . . . . . . . . . . . . . . . . . . . . . . . 112
5.3.3 Data collection: AFM imaging . . . . . . . . . . . . . . . . . . 115
5.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5.4.1 Quantitative similarity within and across shed skins . . . . . . 118
5.4.2 Microstructures are retained on preserved samples . . . . . . . 122
5.4.3 A third independent origin of pitted microstructure in sidewind-
ing snakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
5.5 Discussion and conclusions . . . . . . . . . . . . . . . . . . . . . . . . 128
5.S Supplementary Material . . . . . . . . . . . . . . . . . . . . . . . . . 131
6 Conclusion 146
6.1 Macroscopic whole-body mechanics . . . . . . . . . . . . . . . . . . . 147
6.1.1 Broader impacts . . . . . . . . . . . . . . . . . . . . . . . . . 147
6.1.2 Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
6.2 Microstructural effects on behavior . . . . . . . . . . . . . . . . . . . 150
6.2.1 Broader impacts . . . . . . . . . . . . . . . . . . . . . . . . . 151
6.2.2 Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
6.3 Connecting scales – arboreal microstructures . . . . . . . . . . . . . . 155
A Appendix 162
A.1 Climbing experiment data collection and processing . . . . . . . . . . 162
A.1.1 Optitrack data collection and processing . . . . . . . . . . . . 162
A.1.2 Force data collection and processing . . . . . . . . . . . . . . . 164
A.1.3 Force autocorrelation . . . . . . . . . . . . . . . . . . . . . . . 164
A.2 AFM data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
A.2.1 AFM operation . . . . . . . . . . . . . . . . . . . . . . . . . . 165
A.2.2 Gaussian filter for AFM images . . . . . . . . . . . . . . . . . 167
A.3 Misc. projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
A.3.1 Dorsal microstructures . . . . . . . . . . . . . . . . . . . . . . 168
A.3.2 Glass lizard microstructures . . . . . . . . . . . . . . . . . . . 169
A.3.3 Surface characterization . . . . . . . . . . . . . . . . . . . . . 169
Bibliography 174
About this Dissertation
| School | |
|---|---|
| Department | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Keyword | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Balancing Acts: Multi-level physical interactions provide insights into snake climbing behavior. (Scaling the scales of scaling scales) () | 2026-07-10 11:55:03 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Ch.4 S2 (Downward climbing with 2.5–cm posts spaced 50 cm apart. Video is 5x speed.) | 2026-07-10 11:55:13 -0400 |
|
|
|
Ch.4 S4 (Downward climbing with 2.5–cm posts spaced 100 cm apart. Video is 5x speed.) | 2026-07-10 11:55:24 -0400 |
|
|
|
Ch.4 S5 (Upward climbing with 0.3–cm posts spaced 50 cm apart. Video is 5x speed.) | 2026-07-10 11:55:38 -0400 |
|
|
|
Ch.4 S6 (Downward climbing with 0.3–cm posts spaced 50 cm apart. Video is 5x speed.) | 2026-07-10 11:55:53 -0400 |
|
|
|
Ch.4 S7 (Upward climbing from our robotic model. The force sensitive post is colored cyan in the animation. Video is 5x speed.) | 2026-07-10 11:56:11 -0400 |
|
|
|
Ch.4 S8 (Downward climbing from our robotic model. Video is 5x speed) | 2026-07-10 11:56:29 -0400 |
|
|
|
Ch.4 S9 (Highlighting an instance in movie S5 where the middle of the body changes sides of the post demonstrating active readjustment in the mid-body. Video is 5x speed.) | 2026-07-10 11:56:45 -0400 |
|
|
|
Ch.4 S10 (Brooks’ Kingnakes (Lampropeltis getula brooksi) perform poorly and fail on our climbing wall. Clips are real speed.) | 2026-07-10 11:57:03 -0400 |
|
|
|
Ch.4 S1 (Upward climbing with 2.5–cm posts spaced 50 cm apart. In all of these movies, the black and white video on the left is a video of the experiment where we track reflective markers placed on the back of the snake. The animation on the right shows the tracked body of the snake and the arrows represent the force on each post. Dotted lines show a path across the post perpendicular to the snake’s contact location which makes the deflection apparent. Here forces are generally deflected below this line indicating a tangential force opposite what is predicted by friction. Video is 5x speed.) | 2026-07-10 11:57:18 -0400 |
|
|
|
Ch.4 S3 (Upward climbing with 2.5–cm posts spaced 100 cm apart. Video is 5x speed.) | 2026-07-10 11:57:35 -0400 |
|
|
|
Ch.3 S1 (A characteristic trial of a katydid individual using an alternating tripod gait to climb. The individual ID is 15, the incline angle is 60◦, and the sandpaper grit is 120.) | 2026-07-10 11:57:49 -0400 |
|
|
|
Ch.3 S2 (A characteristic trial of a katydid individual using a metachronal gait to climb (i.e., not tripod). The individual ID is 15, the incline angle is 90◦, and the sandpaper grit is 40.) | 2026-07-10 11:58:05 -0400 |
|