Prenatal Mortality In Mammals!

I read Prenatal Mortality In Mammals by F.W. Rogers Brambell, published in 1948. The intention of this article was to "provide a critical examination of the methods of estimating the amount, and of analyzing the distribution, of prenatal mortality in mammals." In this article Brambell discusses the oestrous cycle and embryology of rabbits, to provide important background, then talks about the number of ova ovulated at oestrus, the methods of estimating the total prenatal mortality, the nature of the limitations of the data, and then the data itself: the percentage of ova lost between ovulation and implantation, the loss of embryos after implantation, and the total loss before birth. Finally, he discusses the removal of dead embryos and draws his conclusions.
First, Brambell discusses the oestrous cycle and embryology of rabbits, who in the wild have an intensive breeding season between February and June (tame rabbits breed continuously). He explains pregnancy in rabbits, and the stages of development, especially mentioning corpora lutea, which are visible macroscopically through pregnancy, and implantation sites, which both become important when he discusses how data about prenatal mortality can be obtained.
An important fact which Brambell stresses, is that the number of ova ovulated at oestrus determines the size of the litter, and depends on the species. By definition monotocous species such as humans, horses, cattle, some bats, elephants, and whales, produce one ovum. Polytocous species, on the other hand, produce many ova. These include rodents, insectivores, and pigs. Marsupials have been observed to produce between 20 and 25 ova, while pig litters average around 20, and insectivores tend to have the most, upwards of 40 having been observed. Brambell explains that the number of ova ovulated can be estimated from the number of corpora lutea in ovaries, but errors can arise from a single follicle liberating two or more ova, a follicle giving rise to corpus lutea without liberating an ovum, or counting errors.
The methods of estimating total prenatal mortality are equally flawed. One can count the number corpora lutea of pregnancy in the ovaries of animals who have recently given birth to litters of a known size, but this is not practical in the wild, and requires the sacrifice of the mother and her litter. It also does not distinguish between loss of unfertilized ova from loss of fertilized ova. Another method is "statistical comparison of the mean numbers of corpora lutea in a set of new-born young in a litter determined from different samples of the same population," but this provides no information about the loss of entire litters or as to the stage at which loss occurs in the surviving litters, and great caution must be taken to ensure samples are comparable. A third method is to determine the "proportion of oestrous periods accompanied by copulation which result in live births," but this is not applicable to animals like rabbits that allow copulation at other times, and again gives no information about the stage at which loss occurs. The final method he explains is a comparison of the number of surviving embryos with the number of corpora lutea in animals at any stage of gestation after implantation, which he seems to favor but must be done in conjunction with the other methods.
Brambell is careful to point out the nature and limitations of the data collected before he presents the data itself. With many options for collecting data, but all of them flawed, several different methods have to be employed to collect adequate data.
Finally he explains the results. The loss of ova before implantation in wild rabbits is from 10.2% to 13.0% of those ovulated, of which 1.0% to 3.6% are lost in litters in which none of the ova survive to be implanted. The proportion of litters lost after implantation "is inversely related to the body weight of the mother, as it is to the size of the litter at implantation," he states, which makes sense seeing as the size of the litter is dependent on the body weight of the mother. Brambell states that 35.7% are lost entirely, with 1.0% lost before implantation and 34.7% lost after implantation. 7.6% of ova are lost in surviving litters, 7.0% lost before implantation and 0.6% lost after implantation.
Then he discusses, very briefly, how dead embryos are removed. There are two options: abortion and reabsorption or resorption. Abortion, or the forcible expulsion of the embryo relatively intact, is most common among monotocous species. Reabsorption is when the embryo is gradually autolysed or mummified in situ, and is most common in polytocous species.
Brambell concludes that "it is clear that the loss between ovulation and full term is not less than 43% of ova ovulated, and may be, and indeed probably is, substantially heavier." He states that estimates based on inadequate methods of analysis can be misleading, which seems to be a pretty obvious statement. He also points out that in wild rabbits a greater part of prenatal mortality occurs at the early stages of pregnancy, before, during, and immediately following implantation, and discusses the significance of this.