| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Institute of Ecology, University of Georgia, Athens, Georgia30602
It's these changes in latitudes, changes in attitudes: Nothingremains quite the same. For all of our running, and all of our cunning:If we couldn't laugh, we would all go insane.—JimmyBuffett
Richards himself expresses surprise that this book has remainedpopular and in demand more than ten years since publication of the firstedition in 1986, which was reprinted in 1990 and 1994. The presentrevised second edition retains the strengths that drove sales of theoriginal, but it remains a limited and flawed coverage of the dynamicfield of plant reproductive biology. That said, I hasten to point outthat I have used Richards's treatment of the subject as the basic"textbook" for my own graduate course in Plant ReproductiveEcology and find that it serves the purpose well. Anyone seeking areasonably comprehensive coverage of this subject would be well servedto buy this book. As a "laboratory manual" to supplementRichards's treatment, I use and recommend Kearns and Inouye's (1993) book,Techniques for Pollination Biologists.
General comments and overview
At first I was struck by the apparent dramatic increase in size ofthe new edition: compared to the 1990 version, which is only 23.3 mmthick, the 1997 edition is 31.6 mm! Returning to my original 1986 firstedition, however, I discovered that it also measured 31.6 mm. Amazingly, all three of these versions contain exactly 529pages! This is so, despite the fact that Richards deleted more than 300references and added more than 500 new ones. It does appear to me thatthe format and production quality have been improved in the secondedition, which is printed on "permanent acid-free textpaper."
I find it interesting that the term "breeding system"figures so prominently in the title to the book, yet it is not evendefined in the glossary! We learn on p. 201 that the breeding systemincludes, but is not limited to, whether a species is self-compatible orself-incompatible, but the full "suite of characters which controlthe breeding system, all of which are important, and all of which mustbe considered together" is never explicated. A list of the majorchapter titles is somewhat illuminating in this regard: (2) Sexualtheory in seed plants, (3) Sexual reproduction in flowering plants, (4)Floral diversity and pollination, (5) Pollination biology and gene flow,(6) Multi-allelic self-incompatibility, (7) Heteromorphy, (8) Dicliny,(9) Self-fertilization and inbreeding, and (10) Agamospermy. Comparedto the first edition, one chapter is deleted entirely: Vegetativereproduction. The other major revisions mentioned by Richards in thepreface are expansion of the chapter on sexual theory and that onself-fertilization. Indeed, Chapter 2 (sexual theory) appears to haveincreased by about 10 pages at the expense of Chapter 3 (sexualreproduction). Chapter 9 (self-fertilization), on the other hand, iswithin one page of its former length (i.e., not expanded). In fact, thegreatest increase in size seems to have occurred in Chapter 10(agamospermy), which has grown from 33 to 55 pages!
I hesitate to criticize an author's coverage of a topic, as I amwell aware of what my own biases would be if I were to take on the samedaunting challenge. And I feel that, in general, Richards has done anexcellent job of covering a wide range of phenomena in a diverse arrayof plants. His solid grounding in basic genetics, as well as naturalhistory, has stood him in good stead to deal with this material. Nevertheless, there are a number of sins of omission, as well ascommission, that I will lay out below for the prospective reader/buyerof this book.
I feel that the book is weakened by Richards's decision todelete all of the material on vegetative reproduction. It is a commonand widespread phenomenon that is of great importance to plants. Ifasexual reproduction is to be ignored, then it makes little sense todiscuss the issue of "why be sexual?" (p. 24). I alsoquestion Richards's claim in the preface that "all referencesto plant groups other than the flowering plants have also beenlost." On p. 298 he launches a diatribe against applying sexualterminology developed for vascular plants to bryophytes, but hisarguments are not developed well. Having considered these issues insome detail myself (Wyatt, 1985), Ithink Richards should either explain his opposing views more fully or,more propitiously perhaps, omit them here.
To me, the greatest weakness of the present volume is Richards'saversion to theory and lack of appreciation of the role of modeling inmodern ecological genetics. Richards's lament, voiced throughoutthe book, is that "theoretical and population studies have grownin popularity" (p. 412) to the detriment of more traditionalapproaches such as Mendelian genetics, morphology, and natural history. In his view, "theoretical advances are sterile andnon-substantive" (p. ix). Moreover, his first edition "wasa book based on natural history rather than mathematics, with verylittle modelling included." He has attempted to remedy thissomewhat in the second edition, but, unfortunately, his efforts havefallen short. He frequently uses ESS (evolutionarily stable strategy)arguments without seeming to understand their basic formulation. Hisdiscussion of sexual selection is very confusing and conflates it withgametophytic selection at one point (p. 36). Sometimes, Richards seemsto avoid discussing relevant models or statistical approaches (e.g., notexplaining mixed-mating models, instead referring the reader to Brown(1990) on p. 370). Other topics are avoided entirely (e.g.,"functional gender" is mentioned in passing on p. 25, butnothing of substance is said) or covered very imperfectly (e.g.,"neighborhood is a rather difficult concept..." on p. 187). The insights contributed by phylogenetic approaches are not evenmentioned!
Among the sins of commission committed by Richards in this book, themost glaring are those related to the inclusion of unpublished data andinappropriate rebuttals of the work of others. I have already alludedto the fact that I believe Richards's coverage of agamospermy isoverweighed relative to other topics. Much of this derives from theinclusion of unpublished data by Richards and his graduate students,which has never been subjected to peer review. Besides the extensivedetailed coverage of unpublished data on Taraxacum (e.g., pp.349, 399), there are references and new tables and figures relating toflower temperatures in Crocus (p. 108), orchid pollination (pp.117, 121), fruit predation in Silene (p. 141), and heterostylyand homostyly in Primula (pp. 270, 276, 391). To me, theseseem excessive and inappropriate to include in a textbook. Moreover,Richards's critique of Barrett's (1983) review of heterostyly(p. 284) and of an esoteric point regarding agamospermy by two Swedishworkers from the 1940s (p. 423) would be more appropriately placed inone of his own papers in the primary literature. Not surprisingly, theauthor most cited by Richards is Richards: five or more dissertations byhis graduate students are cited and at least 21 papers on which he isfirst author!
On a number of subjects Richards's views evince uncriticalacceptance of older, mainstream dogmas or, conversely, eccentricrejection of new thinking about a subject. In drawing attention tothis, I do not mean to condemn Richards or profess to know that he is"wrong." Nevertheless, it is important for readers to beaware of these prejudices.
On p. 7 Richards baldly states that "most plants have evolved amixed reproductive strategy." This view derives from Stebbins (1950) and has been challenged byLande and Schemske (1985), among others. In Richards's defense, I point out that he does offer a morebalanced appraisal of this issue in Chapter 9 (self-fertilization). Onp. 12 Richards assumes that gametophytic self-incompatibility is basalwithin the angiosperms, as proposed originally by Whitehouse (1950). Recent molecularevidence, however, accords more closely with the views of Bateman (1952) and others thatself-incompatibility systems have arisen multiple times in variousclades. The utility of pollen–ovule ratios as indicators of plantbreeding systems seems to be accepted at face value (p. 31), despiteserious shortcomings of this approach and questioning even of itsconceptual basis by various authors. Richards states that "explanation that dioecy has arisen from selfing progenitors in responseto selection pressures which favour outcrossing has a long andrespectable history" (p. 300). This seems to imply that new,ecological explanations, which presumably have a short and disreputablehistory, should be rejected out of hand. Yet it is exactly this freshthinking that has breathed new life into a number of well-worn subjects,such as the evolution of dioecy. Finally, Richards's embracing of"agamospecies" (p. 456), which he defines as"microspecies or segregate species of low taxonomic amplitude usedin agamic complexes" (p. 456), may grow out of his long-termstudies of such plants, but his taxonomic views are distinctly in theminority.
More specificcriticisms and nit-picking
There is a serious problem on p. 44 and the following few pages,where the mathematical formulas have been left out entirely. On p. 65and in the glossary, Richards gives the definitions of"multiovulate" and "uniovulate," rather than of"multilocular" and "unilocular," which heintended. On p. 81, it is stated that North American species ofCypripedium produce nectar, but these orchids are actuallynonrewarding (Gill, 1989). Richardsapparently disagrees with Raven's (1972) view that bird flowers arered because that color makes them less apparent to nectar-stealing bees,arguing instead that birds "are very responsive to red" (p.96). Mention should at least be made of the alternative hypothesis. Inthe plate facing p. 244, Fig. 1a is not the temperatePassiflora incarnata, but rather the tropical P.vitifolia. To my knowledge and according to several taxonomictreatments that I consulted, Spartina patens is not dioecious,as Richards claims on p. 299.
Because I work primarily with mosses and milkweeds, I am, of course,especially sensitive to any mistakes regarding these plants. Apparently, Richards's delimitation of the bryophytes on p. 457excludes the hornworts (Anthocerotae). I was surprised to learn on p.130 that asclepiads do not have specialized pollination mechanisms likethe orchids. Indeed, in my view, the milkweed floral mechanism actuallysurpasses the orchids in complexity! Richards concludes that "itis likely that the regular, actinomorphic flowers of the asclepiadspreclude the attraction of many more sophisticated visitors such asbees." This is nonsense, as even a cursory reading of therelevant literature or a short time spent in the field with these plantswill attest. This myth is perpetuated later as Richards refers to the"butterfly-pollinated milkweed Asclepias exaltata"(p. 187), which in fact is more often pollinated effectively by bees, asare most milkweeds (including even butterflyweed, A.tuberosa). In the glossary (p. 466), Richards overlooks thefact that milkweeds, as well as orchids, produce pollinia.
I enjoyed the rare occasions in this volume when Richards, intendedor not, used puns and colorful language to get his point across. Forexample, on p. 446 it is suggested that sporophytic agamospermy intropical fruit trees, like Citrus, "is certainly afruitful field for further investigation." On p. 87 itwas somehow reassuring to learn that pseudocopulatory bees "rarelyejaculate." I have to admit that it was an issue I would neverhave thought to explore!
A few stylistic and organizational issues are worth considering. Ifound Richards's frequent interruption of the narrative withnumbered lists to be very distracting. I am sure that his intent was tobe precise and concise and to facilitate the reader's rapidrediscovery of certain key issues. For me, however, this convention didnot work. The lists were overused and sometimes used to summarizepoints from unpublished manuscripts. The text includes an odd mixtureof British and American spelling and word usage. "Hermaphrodity" is used in preference to"hermaphroditism" throughout. On p. 149 and in the glossaryon p. 463, the word "major" is explained as a North Americanword for "graduate." It actually means to specialize, as ina major field of study, as bees concentrate their foraging on flowers ofa single species of plant. I am also unfamiliar with the use of"self-exclusive" (p. 385) to mean mutually exclusive, ashypotheses can be. Nevertheless, the most foreign aspect ofRichards's style to me is his predilection for using nouns as verbsor adjectives. Hence, we see "resource" (p. 345) and"compartment" (p. 390) used as verbs, whereas"hermaphrodite" is used in preference to"hermaphroditic." Harlan Lewis(1962) will certainly be astounded to learn that his"catastrophic selection as a factor in speciation" (p. 364)has been transformed into "catastrophe speciation." How docatastrophes speciate, anyway?
A final comment is necessary regarding the illustrations in the book. Some are excellent, especially the ones provided by M. C. F. Proctor,but others actually detract from the overall quality of the book. Manyof the black-and-white photographs, such as those ofpurple-and-white-flowered fritillaries on p. 196 or violets on p. 365,do not show what the author intended and could be eliminated to goodeffect.
The bottom-line andbroader view
As stated in the preface to the book that I edited for Chapman andHall (Wyatt, 1992), the field of plantreproductive biology has grown vigorously in the past 20 years and hasbecome increasingly heterogeneous. Evidence of this can be seen by aquick perusal of any recent issue of American Journal ofBotany, whose pages over the past several years have been dominatedby exciting new papers in this field. The task of summarizing all ofthis progress in a thorough and evenhanded way is now Herculean. Richards has made a valiant effort to collect in one place a diverseassortment of otherwise isolated studies. I applaud his effort, whichshould lead to easier identification of new problems forinvestigation.
Here I will comment on some of the developments in plant reproductivebiology that I see as important and which I feel Richards has not givenadequate coverage in his second edition. First, I believe that we areonly beginning to realize the potential value of the rapid growth in ourknowledge of plant phylogenies. Phylogenetic reconstructions at alltaxonomic levels, using new molecular data as well as reinterpretedmorphological data, provide insights into when and where reproductiveinnovations arise. In this way we can dissect out the relevantcharacter transformations to see if they are associated, for example,with a shift in habitat or ecological associates. By mappingreproductive traits onto a phylogenetic tree constructed fromindependent data, we can determine how many times that trait hasevolved, gaining an understanding of the selective forces involved andof how likely it is that parallel evolution can occur. In this regard,the recent work of Armbruster(1997) and Johnson, Linder, andSteiner (1998) is exemplary.
David Lloyd pioneered the application of general selection models tothe evolution of plant reproduction. He and his coworkers haveinfluenced others to use this approach to great effect. Often theresults of these models have helped to guide our thinking in generalabout these issues and to direct our attention to the most importantparameters to measure. Even though they are not necessarily accuratereflections of the real world, they offer insights into how plantsfunction and the likely magnitude of differences that character changeswill make. Following in Lloyd's tradition, Morgan and Schoen (1997) illustrate the utilityof this approach. It is unfortunate that Richards's book fails tocapture the essence of what this approach has to offer.
Another approach that has prospered in plant reproductive biology,but which seems to be downplayed by Richards, is interpretation ofreproductive traits in ecological terms. The original view that alltransitions in breeding systems should be driven by genetic consequenceshas been expanded to accommodate a wider range of driving forces. Forexample, herkogamy and dichogamy, originally interpreted strictly as"outcrossing mechanisms," may more often have been shaped byselection to prevent interference between the male and female functionsof hermaphroditic flowers (Lloyd and Webb,1986). The Darlingtonian hegemony has been broken, withoutthe necessity of rejecting all of its principles. But you would neverrealize this from a reading of Richards's book.
And certainly, I agree with Richards's contention in the preface(p. x) that "evolutionary genetics still underpin all plantbreeding system work, but in almost no case is any geneticsundertaken." It appears that molecular genetics has eclipsedMendelian genetics, often to the detriment of those of us mostinterested in the ecology and evolution of plant reproduction. Usefulinformation has been contributed (e.g., about the existence andstructure of the S-locus in self-incompatible plants), butother basic mysteries that could be explained by classical geneticapproaches have been shunted aside (e.g., if "late-actingself-incompatibility" is indeed under genetic control). We oftenassume a genetic basis to a trait of interest and further assume thatgenetic variation exists on which selection can act. Personally, I amdisappointed at what quantitative genetics has contributed thus far toour understanding of the evolution of plant reproductive characters. The difficult and time-consuming experiments and the complex andambiguous statistical designs are daunting, and the results thusobtained seem so hard to interpret and apply back to the naturalsituation that little seems to be gained for the effort invested. Perhaps Ruth Shaw and others will soon show us a way out of thisconundrum.
Three other technical approaches that have provided valuable newinformation also appear to me to be reaching the limits of theirutility. The ability to use genetic markers in conjunction withstatistical estimators to quantify outcrossing rates was a majorbreakthrough in the early 1980s. It enabled us to compare plantpopulations or species in terms of what was actually achieved withrespect to self- vs. cross-pollination and provided other insights intomating within populations. Attempts to extend this approach to a finerscale, to examine family-level outcrossing rates, have beenproblematical, even in cases where inferences are not limited bysufficient genetic polymorphism. It appears that newer, more robuststatistical methods need to be developed.
The concept of functional gender was envisioned by Lloyd (1980) as a way to avoid classifying plantsinto typological categories and, instead, to quantify the degree towhich they contributed to the next generation as male vs. femaleparents. This approach provided useful insights into the evolution ofdioecy from distyly, but the calculations are tedious for most cosexualplants, depending on data that are technically difficult to obtain. Thus, the approach, though conceptually brilliant, has had less impactempirically than one might have hoped. Nevertheless, it at leastdeserves mention by Richards. On the other hand, paternity analysis hasprovided useful tests of some of our pet ideas regarding gene flow andsexual selection in plants. Again, it is often fraught with analyticaland interpretational problems, even when sufficient genetic polymorphismexists. Many of the statistical "solutions" to the lack ofresolution, such as assigning "partial paternity," areunsatisfying in the extreme.
A few other arenas in which progress appears to be lagging includeour understanding of female choice, male: female allocation, andpollinator effectiveness. Work on female choice seems not to haveprogressed beyond merely demonstrating in a range of plant species thatprogeny of certain crosses are favored on a given maternal plant. Themechanistic basis of this phenomenon remains unexplained. Models thatpredict mating-system evolution often depend critically on measurementof reproductive investment in male vs. female function. Yet we seem tohave made little progress in devising meaningful ways to quantify theseparameters. Many recent authors have discovered that plant-pollinatorinteractions seldom lead to highly coevolved and specializedrelationships (e.g., Waser et al., 1996;Waser, 1998), in strong contrast toearlier ideas on the subject. Given that most plant species aregeneralists, we need better means of measuring pollinator effectiveness. The older tradition of merely listing all plant visitors and guessingwhich are most effective needs to be replaced with a more direct,quantitative assessment. We may then discover that features of certainflowers have been misinterpreted under a too-ready acceptance of thedogma of pollinationsyndromes.
FOOTNOTES
1 Plant breeding systems.
Second edition. A. J. Richards.
Chapman&Hall. 1997. 529 pp. $54.95. ISBN 0-412-57450-0
(Paperback). 
REFERENCES
Armbruster, W.S. 1997 Exaptations link evolution of plant-herbivoreand plant-pollinator interactions: a phylogenetic inquiry. Ecology 78: 1661–1672. [CrossRef][ISI]
Bateman, A.J. 1952 Self-incompatibility systems in angiosperms. I. Theory. Heredity 6: 285–310.[ISI]
Gill, D. E. 1989 Fruitingfailure, pollinator inefficiency, and speciation in orchids. In D. Otte and J. A. Endler [eds.], Speciation andits consequences, 458–481. Sinauer, Sunderland, MA.
Johnson, S. D., H. P. Linder, and K. E.Steiner. 1998 Phylogeny and radiation of pollinationsystems in Disa (Orchidaceae). American Journal ofBotany 85: 402–411.[Abstract]
Kearns, C. A., and D. W.Inouye. 1993 Techniques for pollination biologists.University Press of Colorado, Niwot, CO.
Lande, R. and D. W.Schemske. 1985 The evolution of self-fertilization andinbreeding depression in plants. I. Genetic models. Evolution39: 24–40.[CrossRef][ISI]
Lewis, H. 1962 Catastophicselection as a factor in speciation. Evolution 16:257–271.[CrossRef][ISI]
Lloyd, D. G. 1980 Sexualstrategies in plants. III. A quantitative method for describing thegender of plants. New Zealand Journal of Botany 18:103–108.
———, and C. J.Webb. 1986 The avoidance of interference between thepresentation of pollen and stigmas in angiosperms. I. Dichogamy. New Zealand Journal of Botany 24: 135–162.[ISI]
Morgan, M. T., and D. J.Schoen. 1997 The role of theory in an emerging newplant reproductive biology. Trends in Ecology and Evolution12: 231–234. [CrossRef]
Raven, P. H. 1972 Why are bird-visitedflowers predominantly red? Evolution 26: 674.[CrossRef][ISI]
Stebbins, G.L. 1950 Variation and evolution in plants. ColumbiaUniversity Press, New York, NY.
Waser, N.M. 1998 Pollination, angiosperm speciation, and thenature of species boundaries. Oikos 81:198–201.[CrossRef]
———, L. Chittka, M. V.Price, N. M. Williams, and J.Ollerton. 1996 Generalization in pollination systems,and why it matters. Ecology 77: 1043–1060. [CrossRef]
Whitehouse, H. L.K. 1950 Multiple-allelomorph incompatibility of pollenand style in the evolution of the angiosperms. Annals ofBotany 14: 198–216.
Wyatt, R. 1985 Terminologyfor bryophyte sexuality: toward a unified system. Taxon 34: 420–425. [CrossRef][ISI]
———. 1992 Ecology andevolution of plant reproduction: new approaches. Chapman and Hall, NewYork, NY.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
