Association, of which there is only a seven-page English language summary and there are not enough details in the summary to evaluate the quality of analyses.
Two of the milk composition studies were carried out in New Zealand. Wells et al. (2004) compared milk from six Freisen clones to milk from the non-clone donor and found statistically significant differences in bovine serum albumin and levels of linolenic and linoleic acid. However, the authors concluded that the levels of these nutrients were within the normal range for the breed and so considered the differences not to be meaningful. The same lead author later published a paper a year later that included the six Freisen clones from Wells et al. (2004) and three more Freisen clones and compared the levels of various vitamins, minerals, and amino acids in the milk of these clones to five comparator animals and found no statistically significant differences (Wells, 2005). However, no information is given on the comparator animals, so we don’t know how they were chosen or whether they were kept under the same husbandry conditions and fed the same diets as the clones.
For the bovine meat composition data, FDA looked at only two published studies that involved only 5 cloned (e.g. SCNT) Japanese Black beef cattle. Tian et al. (2004) did a controlled study but it involved only 2 cloned bulls. Interestingly, there were four surviving “apparently normal” cloned bulls, but only 2 of them were used for the study with no discussion as to how those two bulls were chosen. There were two sets of controls: genetically-matched comparator bulls produced using semen of the son of the donor bull and fed the same diet as the clones, and breed-matched bulls which were simply Japanese Black beef cattle of the same age as the clones. Out of the many variables measured, there were 12 where the clones and genetic comparators showed statistically significant differences. For 10 out of 12 of these significant measurements, the clones were higher than either the genetic or breed comparators. The authors considered the differences to be the result of the superior genetics of the donor animal. Takahashi and Ito (2004) looked at the caracass characteristics of a Japanese Black beef cow, but the study involved only a single SCNT clone that was sampled at 28 months of age, so no statistical analysis could be performed
In addition to the two published studies, FDA also looked at meat composition data on 11 cloned cattle and 11 comparators supplied by a private company, Cyagra. This industry study found “no biologically significant differences” in the meat composition data between clones and comparators. However, the study is poorly designed, with animals of widely differing ages being compared, which makes it virtually impossible to find statistically significant differences. Rather than compare animals of roughly the same age reared under the same conditions, Cyagra used six females clones aged 15 to 43 months and five male clones aged 12 to 17 months. The eleven comparator cattle were simple “over 12 months of age.” It is not surprising that such a poorly designed study found “no biologically significant differences.”