The scaling of postcranial muscles in cats (Felidae) I: forelimb, cervical and thoracic muscles
The scaling of postcranial muscles in cats (Felidae) II: hindlimb and lumbosacral muscles
As you can probably guess, they are two closely related papers covering most of the post-cranial muscles of different felids. Surprisingly it has never really been done, with the only data before these papers were published on cat muscles covering cheetahs, which we used. Another paper has come out since my papers were accepted on the Eurasian lynx (Lynx lynx) which would have been great to incorporate and I'm hoping someone builds on my research and incorporates more individuals and more taxa.
So first a bit of background. This was a collaboration with a lot of people - all of team cat (myself, John, Anjali, Marcela and Stephanie Pierce), plus Andrew Kitchener (National Museums Scotland) and Emily Sparkes (the invaluable technician in Structure and Motion in RVC who suffered many hours dissecting with me). We obtained all of the specimens from zoos and private collections when they had died from natural causes. None were killed for the research, and none were wild. This does have implications for the muscles and the condition of the animals, but it's the best we can do. Funnily it was easily to get a snow leopard (we got 2 in 4 months) than it was to get a domestic cat (we only got 2 in the year we were working on it), due to the levels of consent and ethics required for animals that were pets.
After lots of dissections, which in my case were back breaking (in my case almost literally as I ended up with sciatica for a while from all of the dissections), we had collected data on the muscle body (the bit without the tendon attached) mass and lengths, tendon (where there were) mass and lengths, fascicle lengths (roughly the muscle fibre lengths) from black-footed cat, caracal, ocelot, domestic cat, snow leopard, leopard, tiger, lion. We also measured the pennation angle (the angle of the fascicles to the line of action of the muscle), and combined with the fascicle length and muscle mass we can calculate the physiological cross-sectional area (PCSA: an approximation for force production).
As we were testing to see how these metrics scale, it is worth considering the expected relationships. We were scaling everything relative to body weight, so unsurprisingly the isometric (where things scale as expected from the geometry) scaling for tendon and muscles masses is 1 (i.e. if a cat gets twice as heavy, a given muscle or tendon gets twice as heavy). If the metric differs from that statistcally, we say it scales allometrically (either positively if bigger than expected, or negatively if smaller). Moving onto the other measures, we have to consider masses approximate volumes, so mass is proportional to length3, therefore isometry for the length against mass is 1/3 (when both values are logged), and for the PCSA isometry is 2/3 (area being length2).
The results show that most of the muscle metrics that we studied scaled indistinguishably from isometry whether it be forelimbs, hindlimbs or vertebral muscles. Below are the figures that show which muscles scale allometrically (they are all the pretty colours).
Whilst there are lots of colours, the one I am going to focus on are the greens, PCSA (remember this is linked to force production). As isometry for PCSA scales proportional to body mass2/3, if PCSA scales with isometry big animals become relatively weaker. In fact, even with positive allometry, unless the scaling is greater than 1, the biggest species still become relatively weaker than their smaller relatives. It might be quite obvious where I am going with this if you look at the figures, but ultimately the results show that the felids get relatively weaker as they get bigger. Thus for it's size, a domestic cat is relatively stronger (and previous work shows relatively faster) than its bigger relatives. Some interesting patterns do show up for the PCSAs in the forelimb: muscles linked to shoulder stabilisation generally scale with positive allometry, as do those linked to prey capture (whether it is those active whilst gripping prey, or those responsible for the claws being unsheathed).
Beyond just looking at the muscles, we put the data through a principal components analysis (PCA - not to be confused with PCSA). A PCA works by effectively regressing all of the data against the other data simultaneously, so instead of the standard x vs y graph you end up with each axis potentially representing many variables (e.g. a-z vs a-z but to differing contributions on each axis) to see if the data groups. We tested for both body size (big cat vs small cat - see earlier post), and also locomotor mode (those that basically do nothing but stay on the ground vs those that also regularly climb) after removing the effects of size from all of the data to see if there are any differences.
Whilst it appears that there is separation of the groups on some of these groupings, statistically there is no difference between any of the pairings we compared. It may ultimately boil down to the fact that cats are very conservative across all species and their muscles aren't scaling that differently to what we expect, and ultimately their physiology and ecology means they are high speed, ambush predators but for most of their time, they do very little. Is this ultimately how big cats can get away with being relatively weaker than their small relatives?
Cuff et al., 2016a. The scaling of postcranial muscles in cats (Felidae) I: forelimb, cervical and thoracic muscles. Journal of Anatomy DOI: 10.1111/joa.12477
Cuff et al., 2016b. The scaling of postcranial muscles in cats (Felidae) II: hindlimb and lumbosacral muscles. Journal of Anatomy DOI: 10.1111/joa.12474