Put another way, it's all about the money
Not long ago a colleague and I briefly discussed and summarised the most recent evidence regarding fatty acid (membrane) composition and aging. Importantly, mitochondrial fatty acid composition along with reduced mitochondrial ROS production in long-lived species is one of the major pillars of the "mito-free radical theory of aging" derived from comparative studies.
I would like to emphasize a few additional key points. For a broader state of the art review, the reader is refered to the literature (5, 6). Not long ago I noticed a letter by Barja (1), discussing a major headache for anyone doing comparative biology of aging. One way of putting his idea is as follows:
Body-size (B) is correlated with lower metabolic rate (MR), lower extrinsic risk of death (ED) and higher lifespan (LS). Low ED is linked with high lifespans and thus other traits responsible for longevity (LT). Thus as a rule of thumb the higher body size (B), the higher lifespan (LS). But any longevity trait (LT) associated with lifespan (LS) would necessarily be associated with B.
Therfore we can make two statements:
If we adjust for body-size: The association of any true LT with LS will be weakened. Only if we have enough animals with LS variation at the same B, a significant correlation may persist.
If we do not adjust for body-size: Any trait associated with B would appear as if it is a true LT by pure coincidence
It's not that one view is superior to the other. It simply appears this question is not resolvable given our current means as both adjustments introduce unwanted bias, one bias promotes false positives, the other false negatives.
So for instance, when Csiszar et al. claim that (3):
"body size appears to have a major influence on mitochondrial ROS production in primates."
they are describing a potentially coincidental correlation. However, the same goes for Barja who disregards body size completely (1).
Why is it about money?
If we had the resources we need and deserve, we as biogerontologists, could do large well-powered studies from the get go and would not have to quibble about a question that may be unanswerable given the tools at hand.
The solution to the problem is either meta-analysis, pooling individually underpowered studies, or cleverly designed interventional studies (e.g. ref. 2). For meta-analysis, I currently favour phospholipids as a proxy, because their biochemical measurment is extremely easy compared to e.g. mtROS production in isolated mitochondria (classic paper, ref. 4)
As far as comparative studies are concerned the "clever design" will have to involve parameters that are easier to study to allow for very large datasets. A less clever, but still feasible solution would be to fund studies with adequate sample size, perhaps this must be championed by an international team. If I had to speculate based on my reading of the literature, n>200 for "easy to measure" endpoints like mitochondrial phospholipid composition would be helpful. Studies using more complicated markers often have an n=10-30 (e.g. ref. 3, 4) and extending that is necessary as well. Important endpoints for comparative studies include: mtROSp[roduction] in isolated mitochondria; clever measurment for mtROSp (e.g. "exomarkers"  or redox sensitive proteins) mitoSOX and other fluorescence measurment of mtROSp; phospholipid composition; labile iron pool; proteasome and autophagy activity; DNA repair; markers of DNA damage and mutations/deletions (c.f. ref. 5); sensitivity to stressors (esp. glucose and H2O2, c.f. ref. 3) and so on.
1. Correlations with longevity and body size: to correct or not correct?
Gerontol A Biol Sci Med Sci. 2014 Sep;69(9):1096-8. doi: 10.1093/gerona/glu020. Epub 2014 Feb 25.
2. The Influence of Dietary Fat Source on Life Span in Calorie Restricted Mice.
López-Domínguez JA, Ramsey JJ, Tran D, Imai DM, Koehne A, Laing ST, Griffey SM, Kim K, Taylor SL, Hagopian K, Villalba JM, López-Lluch G, Navas P, McDonald RB.J Gerontol A Biol Sci Med Sci. 2014 Oct 13. pii: glu177. [Epub ahead of print] PMID:25313149
3. J Gerontol A Biol Sci Med Sci. 2012 Aug;67(8):841-52. doi: 10.1093/gerona/glr216. Epub 2012 Jan 4. Testing the oxidative stress hypothesis of aging in primate fibroblasts: is there a correlation between species longevity and cellular ROS production?
Csiszar A1, Podlutsky A, Podlutskaya N, Sonntag WE, Merlin SZ, Philipp EE, Doyle K, Davila A, Recchia FA, Ballabh P, Pinto JT, Ungvari Z.
4.Aging Cell. 2007 Oct;6(5):607-18. Epub 2007 Jun 27. Low rates of hydrogen peroxide production by isolated heart mitochondria associate with long maximum lifespan in vertebratehomeotherms.
Lambert AJ1, Boysen HM, Buckingham JA, Yang T, Podlutsky A, Austad SN,Kunz TH, Buffenstein R, Brand MD.
5. Longev Healthspan. 2014 Apr 1;3(1):4. doi: 10.1186/2046-2395-3-4.
A midlife crisis for the mitochondrial free radical theory of aging.Stuart JA1, Maddalena LA, Merilovich M, Robb EL.
6. Longev Healthspan. 2014 May 1;3:6. doi: 10.1186/2046-2395-3-6.eCollection 2014.
Mitochondrial oxidative stress in aging and healthspan.Dai DF1, Chiao YA1, Marcinek DJ2, Szeto HH3, Rabinovitch PS1.
7. Biochim Biophys Acta. 2014 Feb;1840(2):923-30. doi: 10.1016/j.bbagen.2013.05.026. Epub 2013 May 30.Using exomarkers to assess mitochondrial reactive species in vivo.Logan A1, Cochemé HM, Li Pun PB, Apostolova N, Smith RA, Larsen L, Larsen DS, James AM, Fearnley IM, Rogatti S, Prime TA, Finichiu PG, Dare A, Chouchani ET, Pell VR, Methner C, Quin C, McQuaker SJ, Krieg T, Hartley RC, Murphy MP.