I. SCIENTIFIC BACKGROUND

The term cloning has many meanings. Scientific meanings are reasonably clear, although they have become more complex since technologies for reproducing mammals by cloning from nuclei of somatic cells were demonstrated by Keith H. Campbell and colleagues (1996) and Ian Wilmut and colleagues (1997), the latter resulting in Dolly, the first sheep cloned from an adult somatic cell. Since then, there has been an explosion of research in this area, and the terminology has sometimes been controversial. This entry will cover scientific aspects of both reproductive and therapeutic cloning.

Definitions

Etymologically, clone is derived from the Greek word klon (twig). The ancient Greeks already knew that planting a twig from a tree or bush generally resulted in a new organism very similar to the parent tree. Hundreds of species of plants routinely reproduce by cloning, both at the hand of mankind (e.g., potatoes, asparagus) and naturally (e.g., aspen trees). So, what does "reproduction by cloning" mean?

There are two main approaches to biological reproduction: sexual and asexual. In almost all cases, sexual reproduction involves the processes of meiosis and fertilization. Asexual reproduction does not include these processes. For example, seeds are products of meiosis and fertilization by pollen, and planting these embryos results in sexual reproduction. This is fundamentally different from cutting a potato into several pieces and planting them. Thus, cloning can be broadly defined as asexual reproduction.

There is plenty of asexual reproduction in animals, too. If one appropriately bisects a planarian (a flatworm) or various other invertebrates, two normal copies eventually result. The situation becomes less flexible with vertebrates, particularly with mammals. Nevertheless, even in mammals, asexual reproduction occurs when identical twins or triplets (or quadruplets, etc.) are produced. The duplication that occurs when one embryo produces two individuals is asexual reproduction, albeit superimposed on sexual reproduction. The production of identical multiple offspring is the norm in at least two species of armadillos, and probably in several other mammalian species.

Cloning can also be defined as transplantation of a nucleus from a cell (see Figure 1) into an ovum (technically, an oocyte, or egg). To understand this process, a few biological principles will be reviewed. The billions of cells in bodies of animals can be classified into two kinds: somatic cells and germ-line cells. The germ-line cells have an element of immortality; certain early embryonic cells divide to form a lineage of cells that divide to form gametes (sperm or oocytes), which, after fertilization, form embryos of the next generation, and so on ad infinitum unless the species becomes extinct. Except for gametes, all cells in the body are diploid, that is they have two similar copies of genetic material, one copy inherited from the sperm, and one copy from the egg. Whenever cells divide, they first duplicate the genetic material so that each resulting daughter cell remains diploid. However, the cells that will form sperm divide twice after duplicating their genetic material, resulting in four haploid (one copy of genetic material) sperm, rather than two diploid cells; similar divisions occur to form haploid eggs.

With this background, the basic principle of nuclear transplantation is simple enough. Instead of fertilizing the haploid oocyte with a haploid sperm, one removes the chromosomal genetic information from the oocyte and "fertilizes" it with a diploid cell (see Figure 2).

The first mammals produced via nuclear transplantation were derived from nuclei of cells of early embryos (around the sixteen-cell stage) in the 1980s by Steen M. Willadsen in Cambridge, England. With this approach, one makes a number of genetic copies of an embryo, not an animal. This, of course, changed with Dolly, whose "parent" was a somatic cell derived from differentiated adult mammary tissue. Thus, cloning via nuclear transplantation is fundamentally different when using nuclei from embryonic cells than when using nuclei from adult cells, in that there is considerable uncertainty about the phenotype (visible characteristics) that will result from the embryo, whereas there will be more information about what will result with a nucleus taken from an adult animal, or even a newborn.

Cloning is often defined very broadly as simply making a genetic copy (or copying an organism)—sometimes with the implications of making many copies. Sometimes clone is used as a noun to indicate a genetic copy.

How Identical are Clones with Each Other?

Clonal, or asexual, reproduction, in nature results in nearly genetically identical individuals. This includes two categories of genetic identity: between parent and offspring, and among offspring. However, for numerous traits, genetic identity does not result in phenotypic identity, either due to epigenetic effects or to environmental effects. The environmental effects are well known, particularly from human identical-twin studies. Epigenetic effects are defined as effects due to genes that vary from organism to organism due to random chance, and therefore, cannot ever be predicted exactly. Epigenetic effects are less well known than environmental effects, but can be huge for some traits, such as different coat-color patterns among clones or identical twins. There is no genetic instruction specifying the color of each individual hair in animals with hair of different colors, but only genetic instructions for the general pattern of hair color. These instructions provide general guidelines about how melanoblasts, which differentiate into cells termed melanocytes, migrate and invade hair follicles during fetal development, but not an instruction whether or not to invade an individual hair follicle. Melanocytes reside in hair follicles and add packets of melanin to color each hair as it grows. Numerous other epigenetic phenomena occur during embryonic and fetal development such as random X-chromosome inactivation in female mammals, different methylation (addition of a carbon atom plus 3 hydrogen atoms) patterns of cytosines (see below), and lengths of telomeres, which make up the ends of chromosomes.

There also is considerable variability in embryonic development due to chance effects. Richard C. Lewontin has described how these effects interact with genotype, and with epigenetic and environmental effects, in complex ways so as to generate considerable differences among clonal sets.

One other potential source of differences among animals cloned from genetically identical nuclei is cytoplasmic (see Figure 1) inheritance, illustrated most clearly by mitochondria. Mitochondria are small cytoplasmic bodies located in all cells (with hundreds per cell). They have numerous functions, including generation of energy for such life processes as muscular movement. Mitochondria have their own genetic information in the form of small, circular chromosomes. These almost always are inherited exclusively from mother via the oocyte. Different maternal lines have mitochondria of different genetic makeup, so it is the cytoplasm of the oocyte that determines the makeup of the mitochondrial genome, rather than the chromosomes in the nucleus. Thus, when cloning by nuclear transfer, the mitochondrial genetics will differ from clone to clone unless the oocytes are all derived from the same maternal line of females.

Another source of differences among clones is mutations in the DNA in nuclear chromosomes or mitochondria. DNA is composed of only four kinds of building blocks, known as adenine, thymine, guanine, and cytosine, or A, T, G, and C, respectively. The genetic makeup (DNA) of the nucleus of each mammalian diploid cell has around 12 billion of these building blocks, theoretically hooked together in precisely the same way when DNA is replicated, so that each daughter cell produced has the same genetic makeup, or order of the four building blocks as the "parent" cell that divided. As one might imagine, there is an occasional error when assembling 12 billion items in a specific sequence, and these errors are one source of mutations. Other causes of mutations include background radiation (with which we are constantly bombarded) and chemical reactions, such as peroxidation, which is a chemical process caused by oxygen that can be very detrimental.

The human body is loaded with antioxidants to prevent peroxidation, and its cells contain DNA proofreading and repair enzymes, but these are imperfect at preventing mutations. A common example of mutations is cancer cells, which no longer have true copies of the DNA of normal cells. Most mutations do not cause cancer or have any other noticeable effect, but some cause changes—such as blue rather than brown eyes. Differences among otherwise genetically identical clones due to mutations are usually minor, but nevertheless do occur frequently.

The "gold standard" for genetic identity of mammals is identical twins, triplets, etc. These at least start out with identical chromosomal and mitochondrial genetics and are gestated in the same environment. Even postnatally, identical twins usually grow up in a very similar environment. All man-made clones will be less identical than these, especially in phenotype. Since there are considerable differences between naturally occurring identical twins, such differences will also occur among manufactured clones, in addition to the other differences already discussed.

Procedures for Cloning Mammals

There are numerous procedures for cloning mammals, but two are the most common. The first concerns making identical copies of embryos from embryonic cells, and the

FIGURE 1

second creates embryos (with identical nuclear DNA) from cells of embryos, fetuses, young animals, or adult cells.

Conceptually, the simplest approach would be to separate the two cells of a two-cell embryo so that two identical organisms form. This has been done repeatedly in one way or another, even occasionally resulting in identical quadruplets when dividing a four-cell embryo four ways. Success rates are quite high when aiming at identical twins, but become very low when dividing embryos into quadruplets, the practical limit of the technique. For technical reasons, this approach is much more practical at later stages of embryonic development—at the 100-cell stage, for example), when embryos can be bisected. This latter approach has been used to produce thousands of identical twins (and occasionally triplets) commercially, primarily with cattle (as illustrated by Timothy Williams and colleagues [1984]).

Surprisingly, the main reason for splitting embryos to produce demi-embryos is not to produce sets of identical copies, but rather because splitting embryos augments the general technology of embryo transfer, which is designed to increase the reproductive rates of agricultural (and other) females, much like artificial insemination increases the reproduction of males. To illustrate, pregnancy rates for whole bovine embryos are around 65 percent, whereas pregnancy rates for half embryos are around 50 percent. Thus, because there are twice as many demi-embryos after the splitting process, the net pregnancy rate is frequently over 100 percent. Identical twins and triplets produced by these methods make excellent experimental subjects because genetic variation can be controlled, and sometimes they are produced mainly for these purposes.

FIGURE 2

With nuclear transfer, the main principle is that the ovum, or oocyte is a minifactory designed to produce an embryo, which eventually develops into a term pregnancy. Half of the genetic instructions to make the conceptus normally come from the oocyte, and half from the sperm. With cloning, a complete set of genetic instructions is provided by the nucleus of one embryonic or somatic cell. Of course, those instructions originally were derived from the sperm and oocyte that resulted in the organism that provided the donor cell.

One problem is obtaining oocytes to use as recipients for the diploid nuclei. These cells, the largest in the body (about 1/200 inch in diameter), must be of the same species as the donor nucleus. Usually, they are aspirated from ovarian follicles (large blister-like, fluid-filled structures). In the case of farm animals, oocytes are often obtained from ovaries of slaughtered animals of unknown background. An alternative is to aspirate (remove by suction) oocytes through a large needle inserted into the ovaries in the body cavity of living animals—ultrasound is usually used to visualize the follicles so the needle can be guided into them after piercing the wall of the vagina. This method is used in women to obtain oocytes for routine in vitro fertilization. Oocytes from laboratory animals such as mice are usually obtained after the oocytes are ovulated (released from the follicles) naturally. The oocytes then are located in the part of the reproductive system called the oviduct, and the body cavity needs to be opened to get them out, either via surgery with anesthesia, or after euthanizing the animal.

After oocytes are obtained, they are cultured under specific conditions with specific chemicals until they have matured appropriately. The length of the maturation period may range from less than an hour to two days, depending on the species, the treatments, and the reproductive status of the animal providing the oocytes.

The next step is to remove or destroy the unwanted chromosomes of the oocyte. This usually is done by aspiration of this material with a micropipette (see Figure 2), although there are other options, such as destroying the chromosomes with a laser. Following this step comes transplantation of the nucleus. This can be done by removing the nucleus from the donor cell and injecting it into the cytoplasm of the oocyte. However, in the vast majority of cases the entire donor cell is simply fused with the oocyte using an electric pulse. This incorporates the nucleus into the oocyte, but it also mixes the cytoplasm of the two cells, which also mixes the mitochondria. This is usually not a problem because the oocyte has more than 100 times the volume of the cytoplasm of the donor cell, so the donor cytoplasm essentially gets diluted out.

When a sperm fertilizes an oocyte, it not only adds its 50 percent contribution of genetic material, it also activates, or turns on, the oocyte. Prior to fertilization the oocyte is a large, slowly dying cell. The sperm adds a specific enzyme that chemically activates the ovum, so it comes to life, starts using more energy, and, among other things, duplicates the genetic material in preparation for division to the two-cell stage. This activation function must be duplicated during the nuclear transplantation process for successful embryonic development. It is accomplished in a variety of ways, depending on the species and other details, such as the degree of maturity of the oocyte. A common approach is to apply a strong electrical shock.

The final step is to allow the cloned embryo to develop in vitro, eventually growing from the two-cell stage to a suitable stage for transferring the embryo back to the reproductive tract of a recipient. The length of this culture is usually a few days to a week, depending on the species.

Potential Applications of Cloning Nonhuman Animals

Aside from splitting embryos to produce more offspring, the main application of cloning to date has been to obtain basic biological information that can be applied in other areas. This will continue to be the main value of cloning for some time, and will result in information about causes of birth defects, aging, cancer, and other disease states.

One obvious application of cloning by nuclear transplantation and cell fusion is to make genetic copies of outstanding agricultural animals. As discussed earlier, a genetic copy does not equal a phenotypic copy, so this is not nearly as attractive as most people surmise. For example, the genetic contribution (heritability) to differences between cattle (within breeds) in milk production is on the order of 30 percent, while other factors, mainly environment and random chance, explain the other 70 percent. Thus, if one cloned a cow producing 3, 000 gallons of milk annually, selected from a herd averaging 2, 000 gallons of milk, on the average only 30 percent of the difference between the production of the individual cow and the herd would show up in the clone. A herd of such clones might average 2, 300 gallons of milk, a substantial improvement over the 2, 000 gallons average, but not even close to the 3, 000 gallons produced by the animal being cloned. (This example is an oversimplification, for a variety of reasons—including interactions between genotype and environment [see Lewontin]— but the broad idea is correct.)

There is an even more serious problem with using cloning to increase production of milk (or meat, fiber, etc.), which is that it is not economically viable. The value of the extra milk produced by such a cow would be less than $1, 000 during her lifetime, and she might eat more feed than other cows because more nutrients are required to make more milk, further decreasing her economic value. Costs of cloning in 2003 are in excess of $10, 000 per cow, and while this likely will decrease markedly, it is unlikely that costs will approach economic viability in the foreseeable future. Thus, herds of cloned cows are not likely any time soon. The situation for meat production is even less favorable economically. If one did use this strategy, there would be hundreds of different donor cows cloned due to wanting different optimal genotypes for different environments (e.g. the optimal Vermont cow would be different from the optimal cow for Georgia) not to mention the individual preferences of farmers).

One agricultural application that does make sense is to make copies of genetically (as opposed to phenotypically) outstanding individuals. A good example is a bull whose daughters, on the average, have excellent milk production and are not prone to mammary gland infections. Such a bull might have thousands of daughters demonstrated to be superior to the average population. This bull obviously is essentially worthless phenotypically—copies will not produce any milk—but cloned copies of the bull will produce essentially identical sperm that can be used to produce more daughters by artificial insemination. For this example, one or two clones would likely produce all the semen that could be sold, so large numbers of copies are not needed. In fact, the main application in this context is insurance. Such bulls are extremely valuable, and having one or two copies makes good economic sense. More copies, however, are redundant and expensive to feed and maintain.

Another popular potential application of somatic-cell cloning concerns companion animals, particularly dogs and horses. Again, one will not get a phenotypic copy, so this only makes marginal sense. The resulting cloned animal will often have somewhat similar coat-color patterns and be roughly the same size, but it may have a very different personality, since this is largely influenced by environment. One does not recreate the same animal by cloning, simply a chromosomal genetic copy.

There are myriad experimental uses of cloning, particularly in making transgenic technology more useful. Cloning by nuclear transplantation is thus a powerful experimental tool.

Potential Applications of Human Cloning

In most cultures there would be huge ethical problems in making genetic copies of human beings—so-called reproductive cloning. Currently, this is ethically unacceptable because of the high incidence of congenital abnormalities in offspring derived from cloning by nuclear transfer. If there were no such problems—if cloned children would be as healthy as those produced naturally—one can concoct scenarios for which reproductive cloning might be ethically acceptable. The classic example is a couple whose baby dies within a day or two of birth due to an accident that also makes the mother incapable of reproducing due to damage to ovaries. One could theoretically take cells from the dead baby and clone them using a donated oocyte, which could then be transferred to the uterus (which is still functional) of the woman. The donor cells from the dead baby could also be frozen for later use, so timing would not be a problem.

Other (very improbable) scenarios could be envisioned that would make reproductive cloning ethically acceptable for most people. In any case, this technology for reproductive cloning of persons would likely work with a similar success rate as occurs in other species (extremely low, as of2003). It is certainly possible that a century or more in the future this mode of reproduction will be used to some extent, and persons from that era may well consider our current collective thinking quaint. Since chromosomal genetic identity never results in phenotypic identity, one never recreates a person or animal, and even if phenotypic identity were possible, such individuals would still be individuals. Identical twins and triplets provide some guidance on potential problems. Such individuals usually lead fairly normal lives, and they are considerably more identical than manufactured clones will ever be.

Therapeutic Cloning

A second kind of cloning, therapeutic cloning, is intended to produce tissue and organ replacement parts. There are millions of people worldwide who suffer from debilitating diseases such as diabetes, heart disease, and cirrhosis of the liver. Similarly, millions suffer from accidents that severely damage tissues and organs, including burns, spinal cord damage, and crushed kidneys. In many of these cases, tissue or organ transplants will prolong life and greatly increase quality of life. There are two major problems with this approach: (1) There is a critical shortage of such tissues and organs, and (2) there is usually immunological incompatibility of donor and recipient, which requires immunosuppressive therapy that is debilitating and greatly increases the incidence of cancer.

A solution to this unfortunate situation is to use nuclei of somatic cells of the subject to make immunologically compatible tissues for replacement parts. This approach is not yet available for practical use, but likely will be developed in one form or another in the near future. What is envisioned is to take cells (e.g., from skin) of the person who needs the replacement tissue, and fuse them with donated oocytes from which original chromosomes are removed to form early embryos. Instead of transferring these to the uterus to form a fetus, they would be induced to develop into various tissues in vitro . No fetus would be formed, so there would be no brain, heart, leg, or face, but rather tissues that make up body parts. Quite a bit is known about how to induce the embryonic cells to make muscle, skin, or other tissues, but there is still much to be learned.

This approach likely cannot be used to produce a heart or a kidney, at least in the foreseeable future, but producing heart-muscle cells, nerve cells, pancreatic tissue, liver tissue, or skin does seem feasible. Liver, for example, has a remarkable regenerative capability, so only a small bit of liver may be needed—such as liver stem cells, which might regenerate a whole organ after transplantation. Producing pancreatic tissue to alleviate diabetes would likely be considerably simpler, while producing nerve cells to repair spinal cord damage would likely be more difficult.

It is possible that some tissues can be generated from adult stem cells, circumventing the need for cloning via embryos. However, the embryonic approach has several theoretical advantages—it is the way tissues develop naturally, for example—and it has some practical advantages as well. Furthermore, research into in vitro differentiation of tissue, much of which can be done in animal models with or without the cloning steps, will likely produce information that can eventually be used outside of the context of cloning to accomplish the numerous therapeutic objectives.

Characteristics of Cloned Animals and Related Ethical Consequences

If all goes well, a genetic copy of the animal being cloned is produced, but, again, one clone can vary considerably in phenotype from the donor for numerous traits. Unfortunately, natural reproduction does not go well in every case, and such problems are greatly exacerbated with cloning. In a 2002 summary of all available information on animals cloned from somatic cells (38 studies resulting in 335 subjects in 5 species), Jose B. Cibelli and colleagues found that 77 percent of the resulting animals were normal, while 23 percent were not. The normal subjects, though mostly adults, had not yet lived out their normal life spans, so additional problems (over and above those due to normal aging) could yet develop. Cloning from somatic cells has not resulted in monsters, but, in most cases, reasonably normal individuals.

However, 23 percent abnormalities, mostly neonatal death, is completely unacceptable ethically for producing children, and for most scientists working in this area that ends the ethical debate on human reproductive cloning. In the Cibelli survey it was noted that many of the animals produced represented the initial, or at least early, studies on cloning in respective laboratories, and that the incidence of abnormalities likely would decrease with more experience and improved techniques. This is already being borne out in the scientific literature, but it likely will be many years before the incidence of problems with somatic-cell cloning will decrease to acceptable levels for reproductive cloning of people. However, this ethical crutch will also likely disappear with time.

A complex ethical question is where to set the boundaries on acceptable levels of abnormalities. Interestingly, a 2002 study by Michèle Hansen and colleagues that looked at children produced via in vitro fertilization showed that congenital abnormalities were approximately double the 4 percent seen with natural reproduction. Most of these abnormalities were not extremely serious and could be circumvented or repaired. Nevertheless, the abnormalities were doubled, and some were serious. Thus, this ethical problem is already with us.

The question boils down to the right of people to reproduce given an increased risk of an abnormal child. Of course, these questions arise outside of the context of assisted reproductive technology, such as the increased risk of a child with Down's syndrome when older women reproduce. Modern science can minimize such suffering (e.g., by genotyping embryos before transfer back to the uterus, and eliminating those that will result in severely abnormal individuals). Another reality is that, in one sense or another, nearly all persons are abnormal. For example, essentially all humans have lethal or severely debilitating recessive alleles in their genetic makeup, which, if matched with another such allele in a gamete of a mate, will result in death of the conceptus or resulting child.

A frequent abnormality that occurs with cloning by nuclear transfer via embryonic or somatic donor cells is fetal overgrowth. It is not unusual for offspring to be 30 or 40 percent larger than normal at birth. In some studies, up to 30 percent of offspring have this condition, known as large-offspring syndrome, and some animals cloned from the same donor are large, some are normal, and some are small— which elegantly illustrates that identical chromosomal identity does not equal identical phenotype. Large-offspring syndrome is not a genetic trait, in that this problem is not transmitted to the next generation when the cloned animals reproduce naturally. Also, Michael Wilson and colleagues showed in 1995 that these excessively large neonates develop into only slightly larger adults. The scientific consensus is that large-offpsring syndrome can be summarized as a genetically normal fetus in an epigenetically abnormal placenta. That is, the placenta from cloned pregnancies is often abnormal, resulting in secondary problems in the fetus that largely correct themselves after birth. Unfortunately, with routine husbandry, the newborns often die because of being debilitated from gestating in an abnormal placenta. Fortunately, with a few days of intensive care starting at birth, such offspring survive reasonably well and develop normally, as shown by Frank B. Garry and colleagues in 1995.

As with human babies, animal offspring derived from in vitro fertilization or long-term in vitro culture of embryos have a much higher incidence of abnormalities than with normal reproduction, but a lower incidence than with cloning (see Kelley Tamashiro and colleagues). Clearly, some (but not all) in vitro manipulations, particularly when the in vitro period exceeds several days, lead to increased problems in resulting offspring. Thus, there is a baseline of problems with natural reproduction, which increases with the amount of in vitro manipulation (and reaches a higher level with somatic-cell cloning). It is likely that these problems will decrease or be circumvented with improved techniques, and also that the basic information obtained will be useful in decreasing birth defects and neonatal problems that occur with natural reproduction.

There are some special problems with a small percentage of pregnancies from somatic-cell cloning that are not just an increase in incidence of naturally occurring problems. In some cases, the immune system appears to be severely compromised, and there can be major problems with the heart, blood vessels, and kidneys that are extremely rare with normal reproduction. Furthermore, there is an unusual amount of embryonic death and fetal absorption or abortion with cloned pregnancies—over 80 percent embryonic and fetal attrition is not unusual (compared with around 30 percent with normal reproduction). Thus, the incidence of problem conceptuses is very high, and most of these die in early pregnancy. This is still another reason that, as practiced at the beginning of the twenty-first century, reproductive cloning should not be done with human embryos.

A final point is that cloning via nuclei from somatic cells is very inefficient, currently on the order of 2 percent success per oocyte. This is due to the multiplicative attrition (or success) of the various steps. For example, if there is 90 percent successful fusion of donor cell and oocyte, with 50 percent dividing into embryos suitable for transfer to recipients, 30 percent embryonic survival until pregnancy can be diagnosed, 20 percent of diagnosed pregnancies developing to term, and 85 percent surviving the neonatal period, the result is an overall success rate of around 2 percent. These are typical current values, and are one reason why the costs of cloning are so high. While success rates are improving, it will likely be some years until overall success even approaches 10 percent. For human reproductive cloning, dozens of women would need to be involved as donors of oocytes and recipients of embryos to produce even one baby—assuming the procedures worked as well as they do with animal models, which is unlikely. This illustrates another ethical issue, in that undue use of scarce and expensive medical resources would be required for clonal human reproduction.

Conclusion

The most important conclusions from this scientific overview are that, although cloning procedures for mammals are yielding huge amounts of important scientific information, current procedures are extremely inefficient and result in a high incidence of abnormalities in offspring. These problems severely limit immediate prospects for applications of cloning mammals due to both financial and ethical considerations. Furthermore, cloning does not and will not lead to reincarnation of an animal or person, but rather to a new individual with considerable phenotypic differences from the genetic donor.

george e. seidel jr.

SEE ALSO: Christianity, Bioethics in; Embryo and Fetus; Harm; Reproductive Technologies; Research Policy; Technology; and other Cloning subentries

BIBLIOGRAPHY

Campbell, Keith H.; McWhir, James; Ritchie, W. A.; et al. 1996. "Sheep Cloned by Nuclear Transfer from a Cultured Cell Line." Nature 380(6569): 64–66.

Cibelli, Jose B.; Campbell, Keith H.; Seidel, George E., Jr.; et al. 2002. "The Health Profile of Cloned Animals." Nature Biotechnology 20(1): 13–14.

Garry, Frank B.; Adams, Ragin; McCann, Joseph P.; et al. 1996. "Postnatal Characteristics of Calves Produced by Nuclear Transfer Cloning." Theriogenology 45(1): 141–152.

Hansen, Michèle; Kurinczuk, Jennifer J.; Bower, Carol; et al. 2002. "The Risk of Major Birth Defects after Intracytoplasmic Sperm Injection and In Vitro Fertilization." New England Journal of Medicine 346(10): 725–730.

Lewontin, Richard C. 2000. "Cloning and the Fallacy of Biology Determinism." In Human Cloning, ed. Barbara MacKinnon. Urbana: University of Illinois Press.

Seidel, George E., Jr. 2002. "Genetic and Phenotypic Similarity among Members of Mammalian Clonal Sets." In Principles of Cloning, ed. Jose Cibelli, Robert P. Lanza, Keith H. S. Campbell, et al. San Diego, CA: Academic Press.

Tamashiro, Kelley L. K.; Wakayama, Toruhiko; Akutsu, Hidenori; et al. 2002. "Cloned Mice Have an Obese Phenotype Not Transmitted to Their Offspring." Nature Medicine 8(3): 262–267.

Willadsen, Steen M. 1986. "Nuclear Transplantation in Sheep Embryos." Nature 20(6057): 63–65.

Williams, Timothy J.; Elsden, R. Peter; and Seidel, George E., Jr. 1984. "Pregnancy Rates with Bisected Bovine Embryos." Theriogenology 22(5): 521–531.

Wilmut, Ian; Schnieke, Angelika E.; McWhir, James; et al. 1997. "Viable Offspring Derived from Fetal and Adult Mammalian Cells." Nature 385(6619): 810–813.

Wilson, J. Michael; Williams, J. D.; Bondioli, Kenneth R.; et al. 1995. "Comparison of Birth Weight and Growth Characteristics of Bovine Calves Produced by Nuclear Transfer (Cloning), Embryo Transfer, and Natural Mating." Animal Reproduction Science 38(2): 73–83.