B. Lukács

President of the Matter Evolution Subcommittee

of the

Geonomy Scientific Committee

of the

Hungarian Academy of Sciences

CRIP RMKI, H-1525 Bp. 114. Pf. 49., Budapest, Hungary

lukacs@rmki.kfki.hu

Is it true that "Frumentum Neapolitanum est gramen caprarum tritico equo mixtum", or there is some confusion in taxonomy? What is really Genus Triticum? Are wheats simply the members of Genus Triticum?

The present years are the Age of Scientific Nutrition. Lot of people (mainly women) are discussing, what should be eaten. E.g. lards vs. vegetable oils. While the two groups are rather similar, it is true that the carbon chains of the second group contain more double bounds than those of the first. Then supporters of oils tell that because of the double valence bonds they are in less danger of heart attack & such. Maybe; but vegetable oils easily burn in the pan, so they are in enhanced danger of cancer.

Similarly: sugar or honey? OK, honey contains vitamines and such, while industrial sugar is a pure chemical molecule C₁₂H₂₂O₁₁. Also, industrial sugar is a dimer of two sugars of better properties. On the other hand, industrial sugar is much cheaper and easy to handle.

And then there are the gruel, paste and bread materials. Here I remain at the European/Near East plants wheat, rye, barley & oat (and in this sequence of details).

In Central Europe wheat bread is the standard bread, but rye bread is also common, and more and more abundant northward. Barley bread is exceptional. But preferences depend on geography. Note that the old English term "corn" primarily means not the American Zea mays, but simply "staple seeds of Poaceae". So in Scotland "corn" is oat.

Now, rye breads do not differ too much, but wheat breads are rather various, both for the raw seeds and for techniques of bread-making. In Hungary the commonest bread is made from the Hungarian variants of the so called Triticum aestivum (2n=42) seeds; the flour is "white", so the outer shell is mainly removed in the mill. This bread has lot of advantages compared to others, e.g.:

Triticum aestivum production/hectare is the highest of all wheats, so the product is cheap.

The grinding technique is good for the teeth.

The bread will be "high", spongeous, soft.

On the other hand, the removed shells contain vitamines, fibers &c., and sometimes more compact bread is preferred.

Now, in the last years some people in Hungary prefer the bread of the Traditional German Bread. I do not know if it has a proper English name at all, the old Linnéan name was Triticum spelta, but now it is not considered a separate species. The texture of the bread is more compact and the colour is light brown. These differences are, very probably, technical. But the seed itself must be a separate subspecies because of the differences in the ear structure.

Now, is it possible that some consumers feel even this subspecies difference? If they feel, is it so relevant that for them the Traditional German Wheat is good and Hungarian aestivum variants are not?

Let us imagine for a moment that the difference is so big. But then "wheat" loses its meaning, as it well be shown.

First let us see the naive picture, which was almost the same as that of the botany of XIXth century, before the recognition of importance of chromosomes. There are thousands of different grass species (5-10,000), monocots, in the family Poaceae=Gramineae.

4 groups of Poaceae are important staple foods and close kins, in natural sequence

wheat > rye > barley > oat

All in respective genuses as Triticum, Secale, Hordeum & Avena, with lots of species in each genus. Namely:

aestivum, araraticum, boeoticum, carthlicum, dicoccoides, dicoccon, durum, georgicum, macha, monococcum, polonicum, spelta, sphaerococcum, timopheevi, turanicum, turgidum, urartu, vavilovii.

afghanicum, ancestrale, cereale, dighoricum, kuprijanovi, montanum, segetale, silvestris, vavilovii

agriocrithon, distichon, intermedium, polystichon, spontaneum.

abyssinica, byzantina, chinensis, denudata, fatua, hirtula, nuda, sativa, sterilis, strigosa, vaviloviana.

(Some of these species were produced by Man, some in Nature; but all are species in se, because all have regular two-word Linnéan names. The wild ones are in italics.)

And there is a near relative genus, the Aegilops, the Goatgrass. All of them are wild, goats eat them, but the similarity is clear. (Of course, wild seeds are smaller.)

Nature is great. Science is ever greater. Greatest is Scientist, who, for example, untiringly collects plants and describe them in details. The first two corn types of Neolithic Man were maybe T. monococcum and H. spontaneum. But in primaeval Egypt & Mesopotamia, some 5,000 years ago, the cultivated crops were T. dicoccon and H. polystichon. Secales were weeds of cultivated Triticum, Avenae of Hordeum, and were taken later into cultivation because of some special tolerances, e.g. rye tolerates better cold climate than wheat.

Nice complicated picture. But modern genetics weeded out most of Triticum, Secale, Hordeum & Avena "species" from taxonomy.

Around 1900 genetics started to accept the big difference between Body and Germ. Body is changing with individual life, but the generative part essentially not. Weismann cut away tails of mice in dozens of generations and in the new generations the tails were as big as originally. Sooner or later came the idea that the information for the new generation, carried by single cells of mother & father, are in the chromosomes within the cells.

Now we know that some information is in the maternal cytoplasm too, but in first approximation we may remain at the chromosomes. Lots of microscopic observations showed the following pattern (which I have simplified deliberately):

There are the same number, forms &c. in all the individua of one species (of course, if they carry the fundamental information).

In the body cells the number of chromosomes is even, 2n, in n pairs, which are indistinguishable even in microscopes (except the sex chromosomes, which, for simplicity will be ignored).

In the generative cells there are no pairs but single members of each pair, n.

In some phase of cell division the members of pairs first approach each other, and then the cell takes the two groups of n and n chromosomes to 2 disjoint points. During this process only real pairs pair, so a complete half set goes hither, another complete half set to thither. (And a natural idea is that this happens in such a way that each different chromosome is somehow biochemically different, while members of the same pair are very similar. So members of a pair find themselves "by smell" when necessary.)

There are exceptions in all points; but they are generally errors with bad consequences. The above scheme is the rule and average.

But then some consequences are clear, e.g.:

1) The taste, nutritive ability &c. are determined mainly by the information of the chromosomes.

2) Organisms of different number n of chromosomes cannot belong to the same species.

3) But organisms with the same n also do not belong to the same species if their chromosomes cannot pair with each other. Even in close relatives, where they carry quite similar information.

Quite fast in last century biologists recognised coherent groups of organisms, mainly of plants, quite similar, but with the multiples of the same n. The 2n form was the "normal", called diploid. The haploid n form was either very weak, or died. The 3n triploid one was sometimes quite nice, but always sterile. For animals the 4n tetraploid form was generally unable to grow up, but in plants many times it gave a robust plant, but with less fertility. And so on.

It was easy to understand the phenomenon by using 3 thumb rules.

1) Blueprints are in the chromosomes. But there are errors. With one set of blueprints the organism gets lot of blueprint errors. With four sets there is always 2 or 3 good copies. Also, they are not only blueprints, but they command the organism to produce enzymes &c. So 4 copies mean that the necessary materials are produced "more eagerly".

2) But animals are finely tuned. And double quantity of everything may be wrong.

3) In cell divisions and in the production of generative cells there is the phase when the cell takes apart the pairs and pull them to 2 points. But if there are not pairs but quadruples, wha is the guaranty that exactly 2 of them will go to both places? Some divisions will be erroneous; and even more so in generative cells.

So for decades some cultivators had the trick. A small number in each plant generation is tetraploid. More if we apply the chemical compound colchicine (which, e.g., can be found in the roots of autumn crocus). They collected the tetraploids. Then took their next generation. Fertility was bad, some plants grew badly, but they selected the best ones. And the next generation and the next. In some cases after many generations they got a fertile plant with more robust stature, greater yield & so. This tetraploid plant is practically a new species (the cross with the diploid ancestor is almost unsuccessful, and even if not, in most cases sterile), but biochemically very, very similar. So the fruit of tetraploid pear is bigger than that of the diploid: sells better.

Soon enough something turned out about Triticum. For T. araraticum, boeoticum, monococcum & urartu n=7. But for carthlicum, dicoccoides, dicoccon, durum, georgicum, polonicum, timopheevi, turanicum & turgidum n=14. Finally, for aestivum, macha, spelta, sphaerococcum & vavilovii n=21. So the first group is diploid, the second tetraploid, the third is hexaploid. Maybe early civilizations polyploidised proto-wheat T. boeticum or T. monococcum to get bigger seeds. Then they were ready with the tetraploid in early Sumer & Egypt, 6000 years ago, and then came Greece & Rome with Triticum aestivum, which is the hexaploid. The first author with the idea that emmer is a tetraploid einkorn was Aaronsohn (1909). 3 grades of cultivation/civilization: einkorn, emmer and bread wheat.

And interestingly, there are grades. Europeans/Near Easterners can erect the ladder: einkorn (n=7) is/was not used for anything else than gruel (but it gives a good gruel), you can make thin (wafer) bread from emmer (n=14), but you can make a high, spongeous bread only from n=21. OK, things are not linear. Triple quantity of blueprint produces the best seed for best flour for best bread. Forward for the octoploid wheat!

Hungarians are bread experts from birth. Although decades of planned economy weakened ancient traditions, we still remember the ancient lore. "Hard" flour is not used in most part of First World (and you may believe that exceptional Southern Italy & Greece are simply too near to Southern kuskus), but Hungarians always boasted about the steely hardness of their flour, which in the same time produced the highest possible bread. Something must still be understood.

Neapolitans cultivate Triticum durum because pasta made of Roman Triticum aestivum would dissolve in boiling water. You cannot make maccaroni from n=21 wheat!

Now a Hungarian laughs: you can; but you must put some eggs into the paste first. OK, tells the Neapolitan, but from n=14 I can do it even without eggs. My wheat is harder! This is the reason to call it durum=hard.

Then the Hungarian is really offended. His wheat is really steely! It is, because it contains a lot of aleurone, in the best possible composition: 3/4 gliadine and 1/4 glutene. The aleurone is hard because of the glutene, and the paste can glue together because of the gliadine. Try to make good strudel without Hungarian flour!

Now, Hungarian strudel (rétes) is really something. You make a paste from n=21 flour. Then put on a flat surface and make it really thin. Then put something (apricot, cherry, soft cheese, apple, nuts, poppyseed or as you want) on top of the thin layer, then involve the compound and fry it in the oven.

Everybody can try it. However it is not good if the paste layer is too thick. In addition, then the woman from next door tells that such strudel can be made even by her six year old daughter.

To get very thin layer of pastry without holes you really need glutene for hardness + gliadine for not breaking apart.

I never made strudel. However surely, Hungarian Official Standards know about bread flour (code BL 80), strudel flour (code BFF 55) and durum flour (code DSL). The first 2 are white, the third is yellowish, so our strudel flour is n=21.

I also did not try to cook maccaroni from strudel flour. I can imagine that durum flour survives boiling water even better than strudel flour, but how better?

Aaronsohn, rediscoverer of T. dicoccoides, had a congenial idea that T. dicocoides is bigger, better &c. and has 28 cromosomes because it is a tetraploid T. boeoticum, but the real situation is more complicated. T. dicoccoides does not behave itself as a tetraploid, its chromosomes does not identify each other as parts of quadruplets (7 of them), but as 14 doublets. So geneticists came with the next idea: emmer is an allotetraploid. Two close however different species hybridized, the offspring would have been sterile but an accidental mutation happened (or a local medicine man introduced some juice of autumn crocus), and the offspring has both genomes, both sets of blueprints. The properties of the new plant then are a mixture of those of T. boeoticum and X. Only: the plant is more robust, being a polyploid. Symbolically the genome is (A+B,A+B) where A is the "original"; that of T. monococcum. A and B are similar blueprints, both on 7 chromosomes.

But what is X? There is another wild diploid Triticum, T. urartu. But it was not good as Parent. Then the guess of MacFadden & Sears (1946) was Aegilops bicornis. Later Harlan & al. (1966) suggested Ae. speltoides. This is still the most probable (Dvorák & Zhang 1990), but the speltoides genome is not exactly what is needed into T. dicoccoides. Then what? Either some mutations happened afterwards, or the true parent was a very near kin of Ae. speltoides and we must collect weeds even more.

But then T. durum, the staple corn of Neapolitans, is not quite wheat, but half Aegilops, goat grass! Or, it is more correct and more expressive in Latin: FRUMENTUM NEAPOLITANUM EST GRAMEN CAPRARUM TRITICO EQUO MIXTUM. Quite a good derogatory slogan; but Rome must wait.

OK; and what is Triticum aestivum? Its blueprints include a third set. And it is again different! The traditional notation is D. D is unequivocally identified: Aegilops squarrosa. You can turn to MacFaden & Sears (1946): they crossed and colchicized T. dicoccoides and Ae. squarrosa, and they got something similar to T. spelta (which with this ceased to be a separate species). The artifact gave fertile hybrids with older cultivated n=21 wheats.

So the wheat of Romans, the triticum, Triticum aestivum, is not even half goat grass, but 2/3!

Look, Triticum should be wheat and close relatives. But there areonly two diploid (so original) Triticum species: Triticum boeticum andTriticum urartu. The cultivated n=7 Triticum, monococcum, is simply thecultivated boeticum, so T. boeticum = T. monococcum. Priority goes with thefirst name, but I do not know which one is older. T. urartu is wild.

T. durum/dicoccoides is a hybrid of a Triticum and an Aegilops. Still it is called Triticum, not Tritilops. T. aestivum (+ a lot of names) is a hybrid of one Triticum & 2 Aegilopses. It is still called Triticum. And rightly: T. estivum was triticum, the Roman corn. But then what is the Genus Triticum and what is Genus Aegilops? T. aestivum was called sometimes in Magyar élet, so Life because poorer people lived mainly on bread (however hark'ye: the highest quality bread of that time on world), and now cannot we tell, what was life?

There was an idea to unify the Genuses Triticum & Aegilops; it would solve the linguistic problem, but not in an elegant way. While the giant genius Sears himself preferred this idea in a time (Sears & Morris 1967), it is told that he did not like it in 1990 (Sears 1990). Another suggestion is that of Wang & al. (1997): that Aegilops speltoides should be renamed as Triticum speltoides. Maybe; I will return to this. However then the Neapolitans eat pure wheat, but the Roman corn is one-third goat grass.

We started from taste, nutritional properties & such of "wheat". Now: what is wheat of everyday language?

According to what we learnt above, "wheat" is any crop belonging to Genuses Triticum and Aegilops which are used as raw material of bread, pasta or gruel. And then what have we?

I did not taste the wild species of the genuses; they are not even fodder. Goat is a domestic animal, but it eats bushes and such, not regular and deliberate fodder. But looking at the seeds of Ae. (T.?) speltoides and Ae. squarrosa, they would not give very unfamiliar flours.

Now let us see the cultivated wheats. I do not guarantee to mention all; but I mention some more.

So the first is T. monococcum; I did not eat it and it is told to be good for gruel. If anybody has an interesting experience about, please write me.

T. dicoccon/durum denotes one species, so fundamental biochemical differences are improbable. A lot of us ate T. durum, as maccaroni. I can tell you: Hungarian durum spaghetti and aestivum spaghetti have similar, but significantly different tastes, and not only because eggs white in the second. (I know egg taste.) As for wafer bread, see the pita bread, the fundament of gyros and such.

T. aestivum, the common bread wheat is familiar for lot of us. The taste of this bread is definitely not of pita bread. The crop is the same as T. spelta, the Traditional German Wheat on species level: still Hungarian T. spelta bread is slightly different from T. aestivum bread. Thence comes my guess that because of different grinding technique, giving not white flour.

T. timopheevi, the wheat of Armenian wafer bread. I ate it, and it differs from pita bread. No surprise: its blueprint is (AGAG). But what is G? G is as questionable as B. It is again believed to be close to Ae. speltoides. It is much closer to speltoides than to anything else.

T. zhukovskyi, the wheat of Armenian spongy bread. I ate it too, and its taste is different even from the Armenian wafer bread. It is (AA'GAA'G). And what is A'? I will return to this point, but A and A' are blueprints for different species.

In due course I will continue this line of thinking, but it is better now to see A'.

According to a recent paper (Huang & al. 2002), the A' of T. zhukovskyi is the genome of T. monococcum, while A in T. durum and T. aestivum is from T. urartu. So the wild ancestor of the cultivated diploid wheat was not parent of the tetraploid wheats of Sumer, Egypt and the Mediterranean, but was that of the hexaploid wheat of the Caucasus. Interesting; it is better to reconsider the archaeological consequences.

What was the wild corn eaten by the first harvesters (but not cultivators!) of the Fertile Crescent some 9,000 years ago? We believed that T. boeoticum/monococcum. Then the first cultivators hybridized it, and then... But if T. urartu was not cultivated on diploid level, and T. monococcum was not hybridized to get tetraploid "wheat"?

But it seems that wild (ABAB) do exist (Aaronsohn 1909). Indeed Huang & al. could tell only that the tetraploid T. dicoccoides is not older than 500,000 y. So it is possible that humans found T. dicoccoides and simply took it into cultivation. And the areal of T. boeoticum reaches Eastern Anatolia, so proto-Armenians might have met it to hybridize it into T. timopheevi.

Huang & al. calculates some divergence times. Their guess is < 1 My within Triticum (so between boeoticum/monococcum & urartu), 2.5-4.5 My between Triticum & Aegilops (is speltoides Triticum or Aegilops?), 7 My for Secale from the common ancestor of Triticum & Aegilops, and 11 My for Hordeum from all of them.

And this is the proper moment to go back to breads. We are not yet ready.

The poor relatives of Triticum are not the Aegilops goat grasses;goat grasses are the black sheep. No bread eater wants to remember that he eats in 2/3 part the same as goats eat. For a Southern Central European the poor bread is rye bread. Partly because it is more brown, partly because it is eaten on the North.

Rye is (of course) n=7. Botanists of wider interest call the genome (SS), but it is impossible here, because from Triticum/Aegilops viewpoint S is the genome of the Aegilops section Sitopsis, and most definitely that of Ae. spelta. So here I will call the rye genome (RR). Lot of us are familiar with at least rye bread. Its taste differs from that of any "wheat" breads.

And now comes another (two) genus hybrid(s). Only they have the same name, which is not fortunate.

Triticale 1 (hexaploid). Genome (AABBRR), n=21, and Triticale 2 (AABBDDRR), n=28. I did not eat it yet, but can imagine anything between rye and wheat. The literature not always distinguishes between hexaploid and octoploid triticale. They are rich in lysine. Now, interestingly, Hamilton (***) states that Triticale lacks proper glutene content and so triticale bread is very heavy. One would expect this for the hexaploid but not for the octoploid one. Namely, (AB) does not code a proper glutene content, and (R) neither. But (ABDABD) breads can be leavened, so one might expect that for (ABDRABDR) ones too.

Here I stop. There are reports about many other allopolyploids containing Triticum as a component; but some of them are only animal fodder, and some others were reported from the Soviet Union in the 50's. Now, a part of the latter ones finally did not lead to production, while the others were tricky falsifications in the Lisenko hystery For honest publications you can find a review e.g. in Martinek & al. (2001). As for the other case I give here only two citations about intergeneric hybrids which now are believed frauds. Both report rye seeds in wheat ears and go back to times when First Secretary Stalin (aka Dzhugasvili) kept his eyes on botanical research: Lisenko (1951) and Kretovich & al. (1954). I am afraid a lot of people would lie if absolutely necessary. For any case, we cannot eat what is not produced, independently of the reason.

First let us go back to cca. 1970 and tell the majority opinion (so strong majority that is is pointless to give references) about the zoological superfamily Hominoidea. It contained Familiae Hylobatidae, Pongidae & Hominidae. We are not too interested in gibbons, so let us contrast Pongidae and Hominidae.

Hominidae contained us, compared to animals, but some "borderline animals" as well; namely the genuses Ramapithecus, Australopithecus & Homo. In contrast, Pongidae contained genuses Dryopithecus, Sivapithecus, Propliopithecus, Pongo, Pan, Gorilla & Gigantopithecus. From evolutionary point of view the watershed was between close relatives Sivapithecus and Ramapithecus: Ramapithecus "chose the hominid way". But, interestingly enough, "the most human pongid dentition" belonged to Gigantopithecus.

Now let us restrict ourselves to recent species which we can study in details. Hominidae contained one recent species, Homo sapiens, while Pongidae 4: Pan trogdolites and bonobo (chimpanzee and pigmy chimpanzee), Gorilla gorilla (gorilla) and Pongo pygmaeus (orang utan, i.e. in Malay the forest man). The divergence time of hominids and pongids was guessed bw. 15-25 My.

But then came a lot of molecular biologists (again it is pointless to give references). They tell that Homo diverged from Pan cca. 5-7 My ago; from Gorilla 7-8 My ago, and from Pongo cca. 13 My ago.

Interestingly enough, this sequence is exactly the same as of Weinert(1941). But clearly one cannot decide taxonomy merely on distances fromus. It was not too great a task to determine the 10 distances amongst 5species, but then it has come out that Homo, Pan & Gorilla are nearer toeach other than any one to Pongo.

Then old Familia Pongidae is polyphyletic, so has no divergence time. However new, narrower, Pongidae, containing the only recent species Pongo pygmaeus, but together with the proper set of extinct relatives, is monophyletic, with still a divergence time from us 13 My. "Only" the meaning of we has changed. Perhaps still we want to call us as Hominidae; but then chimps & gorillas are also hominids, so hominidization does not mean the Descent of Man vs. Chimp. Perhaps we will call our family Pan, and Homo will be a genus in this family. (If we differ from chimps on genus level, which is not sure at all.) But surely, we have to tell first, exactly who belong to the two taxa of whose divergence times we speak.

So, which plants are Triticum and which are Aegilops? True, some goats eat Aegilops, goat grass, and do not eat Triticum, wheat. But they do not have such a specific taste: we simply do not permit them to eat Triticum. Try to leave them in a developing T. aestivum field for a time...

Triticum got its genus name from T. aestivum, the triticum of Latins, but that is a three-rooted hybrid. The correct way is to base the genus on diploids first. Then surely the starting point of taxonomy must be T. monococcum/boeoticum and T. urartu. These two species are reported to be near to each other, still good species, and both had important role in the history of "wheat". Now, surely, some Aegilopses will be nearer to the 2 Triticum diploids than to other Aegilops diploids, and then they can be classified into Triticum.

Partial data do exist. A set can be found in Huang & al. (2002): the 2 Triticum species diverged less than 1 My, while Aegilops genus from Triticum some 2.5-4.5 My ago. This time means borderline difference for genuses, but nevermind. The important thing is that we may be in a pongid situation. T. monococcum and some Aegilopses may have diverged 4.5 My ago; but are all Aegilopses so far from T. monococcum?

Dvorák & al. (2003) give all genetic distances within theSitopsis section of Genus Aegilops. Ae. speltoides, the probable source ofgenome B is the farthest member, but this fact in itself is not enough to reclassify it.

Also, Hegde & al. (2002) gives much higher (cca. 5 times) distance between Ae. tauschii/squarrosa, father of "wheat" D genome and Secale cereale, than between Ae. squarrosa and Ae. searsii. So maybe rye is indeed a distinct genus and not one more goat grass. Maybe…

You may ask: rye or wheat? The meaning of rye is definite. But what does it mean: wheat? Wheat breads are still expected chemically more close to each other than to (pure) rye bread, but not too much so.

If you have problems with one "wheat" bread, you can still try many. There is no unique wheat species; and maybe not even a unique wheat genus.

And taxonomy would need more caution with interspecific hybrids.

While composing this text, I recognised some fundamental inconsistencies in the literature about aleurone, glutene and gliadine contents of different "wheats". So do not believe everything. Maybe somebody will clarify the contradictory points on pure natural scientific grounds. While in some sense Wheat is almost Life, scientists should be objective even about Life. And I would expect objectivity about sacred European Wheat in Japanese scientists, at least. (My tribe is European, they eat wheat bread, but for pastry they use rather Zea mays.)

From experimental viewpoint the question is easy to answer. There are A, B, D and G constituents in wheats. It seems that the A genome is insufficient for proper aleurone. G may or may not be enough, because in Armenia both wafer bread and high bread can be found. As for B and D, Greek wafer bread is maybe not accidentally flat. But Southern Italian (AABB) maccaroni flour is told "hard", which property I believed to be connected with aleurone content. But good Hungarian bread flour is (AABBDD), and it is also told hard, steely and springy. Now, if these terms in Southern Italy are not mere endearments, then the dilemma can be decided experimentally. How high, spongy and steely will be a bread produced according to Hungarian standards, but from Southern Italian maccaroni wheat (or any (AABB) wheat)?

Aaronsohn A. 1909: Verh. Zool. Bot. Ges. Wien 59, 485

Dvorák J. & Zhang H-B. 1990: PNAS 87, 9640

Dvorák J. & al. 2003: Genetics 148, 423

Hamilton R. ***: http://roberthamilton.pageout.net/user/www/r/o/roberthamilton/Lecture_14.html

Harlan J. R. 1966: Science 153, 1080

Hegde S. G. & al. 2002: Crop Science 42, 608

Huang S-H. & al. 2002: PNAS 99, 81

Kretovich V. L. & al. 1954: Biohim. zerna 2, 140

Lisenko T. D. 1951: Sbornik 2, 90

MacFadden E. J. & Sears E. R. 1946: 37, 81 & 107

Martinek P. & al. (2001): Acta fytotech. et zootech. 4, 285

Sears E. R. 1990: private communication to R. Riley

Sears E. R. & Morris Rosalind 1967: In Wheat and Wheat Improvements, eds. K. S. Quisenberry & L. P. Reitz, American Society of Agronomy, Madison, p. 19

Wang G-Zh. & al. 1997: PNAS 94, 14570

Weinert H. 1941: Stammegeschichte des Menschheit. Kosmos, Stuttgart

My HomePage, with some other studies, if you are curious.