Hi Bill, Joe and Ben, All,
I was about finished with this piece, when Joe Halinar's contribution arrived in hosta-open. This is helpful! Science is not made by a single individual but by many. There will always be differing opinions. Plant taxonomy is always a work in progress and contributions such as Joe's and Ben's contribute to that progress.
How many hosta species are there? Not an answer but an explanation of the problems involved:
(Copyright 2002: by W. George Schmid - part of an article to be published in The Hosta Journal)
I hope I can explain what is involved in answering this question but you have to bear with me. One can not explain science without scientific terms. For those who are not familiar with botanical scientific terms, get out your dictionary of botany. Perhaps, some of you might want to go back and read a few pages in my book to grasp my reasoning for coming up with the current species list (as published in The Genus Hosta). In this brief let's do away with all the sub-specific ranks such as varieties (varietas) and forms (forma) and consider the species rank only.
The question as to how many hosta species there are comes up all the time. It is also discussed in the literature. To wit, the latest article by Zonneveld and Van Iren (Genome Size and Pollen Variability as Taxonomic Criteria - Application to the genus Hosta) [in Plant Biology 3 (2001) 176-185]. Also contributed have Chung, Liu, Hayes, and others in separate papers. Also there is much yet unpublished work going on at this time and I am involved in some of this work. It is not possible for me to give results until complete data are available.
In my opinion, quantitative DNA content is not the sole criterion for delimiting species. It is simply another measure which contributes to it. Methodologies utilizing molecular biology, among them visualization of chromosomal segments by the banding technique and the quantitative determination of species DNA are very helpful, but they can not be considered the sole determining factors. As Joe pointed out: "You can't just take your DNA content and say one hosta is a species and another isn't just because of the DNA content."
Having researched the genus Hosta for three decades, I firmly believe that one of the missing links in determining the species comprising the genus Hosta is in many cases the required modern field investigation. I have done some of this work but my limited resources were not enough (nor would it have been humanly possible for me) to study all allopatric and sympatric Hosta populations in the natural habitat. Even Fujita's work was limited geographically and his is undoubtedly the most valuable from a population study standpoint. I doubt that Ben has studied all populations extant. Some of the older work was sloppy to say the least and some species were delimited on the basis of one or two specimens (some of which even had garden origin). Thus, the genus Hosta has been woefully neglected in terms of population studies. There is no doubt in my mind that a number of species are in fact interspecific hybrids. There is also evidence of intraspecific variation which may contribute to the replacement of varieties by intraspecific competition. The environmental pressures caused by natural events as well as wholesale human disturbance of formerly geographically isolated breeding populations may have generated new ecotypes in their respective habitats. Clinal variation also occurs in the genus. It is well known that a species may produce different phenotypes without changing their genotype. Also, polyploidization is an important cause for speciation and that fact is documented. Again it is not an exclusive one. For me, it would be valuable to find out if the degree of ploidy in Hosta correlates with macromorphological features, the ecological preferences of a species and geographic distribution. It is impossible to explain all of this factors away with limited molecular biological studies. Notwithstanding these studies (including Ben's) make a valuable contribution.
In the final analysis, this whole matter hinges on what one considers the definition of a plant species. In my opinion, the phenetic concept, the biological species concept and the recognition species concept must be employed to arrive at a workable and practical definition through which we can recognize species without resorting to molecular biology. Obviously, it would be splendid if molecular methodologies would confirm such arrangements.
In taxonomy the most universally applied definition is the phenic definition of a species. It is to this day the basis of plant systematics. In this, a species is a group of individual plants which occupy the same geographic area, i.e., a population, in which the interbreeding, individual plants are identical in most of their principal and important macro- and micromorphological features; and such population is distinguished from related specific groups by clear discontinuities. In Hosta some of these populations are allopatric in nature, meaning they are specialized in their habitat and usually isolated from other populations by geographic barriers (as H. yingeri occurring only on Teakuksan and Sokuksan islands). Unfortunately, most populations in the genus Hosta are sympatric and occupy large land areas in which they will overlap with related specific populations. The breeding mechanism in such populations is panmictic and the sum of all alleles that can be combined by mating is called the gene pool. What makes the investigation of the genus so difficult is that the discontinuities which are usually clearly definable are no longer so in Hosta. Because of this the gene flow is no longer restricted and interspecific gene flow occurs frequently in these sympatric populations. It is difficult for individual (but sympatric) populations to maintain reproductive isolation and thus to stabilize their respective gene pools. As a consequence, reproductive barriers break down and new combinations of genes may result.
In the final analysis, in my book, I found it necessary to adopt a ``legal'' species definition rather than a theoretical/biological/micro-/macro-morphological one, because the wild populations have not been studied to a degree which would allow a scientific a complete determination of the status of existing taxa. The legal definition is based on the rules promulgated by the International Code of Botanical Nomenclature and a list of the taxa which stand at species rank as published in Index Kewensis (go to hostalibrary.com and click on species to see that complete and new/historical list). I am glad that I based my species listing on LEGAL (ICBN) definitions but some folks have repeated my species list without digging into my explanatory text and realizing that I questioned that list extensively even back in 1990 when I wrote that part of my book. I did not want to cause chaos in Hosta nomenclature at a time when no detailed scientific proof was available as to the true nature of the "legal" taxa (species) within the genus. I understand, of course, that just because we have a legal basis for taxa does not mean that they are in fact "true" species. Please realize that my book was published a decade ago and a lot of work has been done since. I maintained Maekawa's (and others) work on a legal basis and pointed that out, although many thought that I agree with Maekawa's classification, because they did not read deeper into my book. Many readers skipped over my assertions that I believe the genus Hosta to have a much more limited number of good species than previously thought. You find some of my explanations on pages 295-296 of my book. At this point in time no one can come up (with any degree of certainty) with a list of Hosta species. I will write further on this subject in the near future as part of my "species ..." series in The Hosta Journal in which I shall examine some of the good and not so good species. Until then ...
Bill, you had some thoughts about testing the appearance of traits in seedling populations. This might work to a degree, but for one, it is difficult to reproduce the habitat of a populations. Also, there is the problem of dealing with diversity in a population. If one could obtain enough individual and representative plants mirroring the gene pool of that population and then cross these plants (after all that is what happens in nature) one might get good results. To self a single plant selected from a population a random may give incorrect (random) results because it represents just one member of a large breeding population.
Bill, as for a list of species, I provided one in 1991 to bring order into Hosta nomenclature, but I will not do it again until I have better data. I can tell you though that it will be shorted than the list published in my book.
BTW, I use the term "legal" because it best describes the process a plant has to go through to become an accepted taxon (species). The process is proscribed in the ICBN and it can be called legal in the sense of adherence to the articles of the Code.
W. George Schmid
----- Original Message -----
Sent: Wednesday, February 27, 2002 3:22 PM
Subject: Re: 23 species or less?
>Thanks very much for your valuable and sound reply. This must be
>about the first I received from anyone of the Hosta society since
>my paper appeared a year ago.
I think a lot of hosta people like the idea of your research with DNA
content in hostas. What they don't like is your refusal to give
details or to respond to criticism of your techniques.
First, I appreciate your research with DNA content and this is
valuable information, BUT it is only ONE piece of information. This
information has to be used in conjunction with other data, such as
geographic distribution, karyotype, chromosome banding, anatomy and
various genetic traits. You can't just take your DNA content and say
one hosta is a species and another isn't just because of the DNA
Second, you have not given us any indication of what the standard
deviation is for you data. If the SD is 1 pg, then your 95%
confidence level is plus or minus 2 pg. Thus, if you say a certain
hosta has 30 pg DNA, then at the 95% confidence level your DNA content
is 30 +/- 2 pg. Thus, the hosta can have a DNA of 28 to 32 pg. If
you have another hosta with 29 pg, it's 95% confidence level is 27 to
31 pg. Since there is so much over lap of the results there is no way
you can say they really have different DNA content.
I've seen some of the flow cytomotery work done by lily hybridizers
and they generally feel good just to be able to seperate diploids from
triploids from tetraploids. I don't know of any research in any other
plant genera where people are claiming that the DNA content can be the
sole criteria for seperating species. Your DNA research is cetainly
valuable, but it has to be used within its limititations.
To sign-off this list, send email to email@example.com with the
message text UNSUBSCRIBE HOSTA-OPEN
Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com