| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2 Vegetable Crops Research Unit, USDA, Agricultural Research Service, Department of Horticulture, University of Wisconsin, 1575 Linden Drive, Madison, Wisconsin 53706-1590 USA; 3 Department of Plant Taxonomy, Agricultural University Wageningen, P.O. Box 8010, 6700 ED Wageningen, The Netherlands; and 4 CSIRO Plant Industry, GPO Box 1600, Canberra ACT, 2601 Australia
Received for publication November 9, 1999. Accepted for publication February 18, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: phenetics • potato • sect. Petota • Solanaceae • Solanum series Longipedicellata • species concepts • taxonomy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) recognized 232 species, partitioned into 21 series. However, nine of these species are members of separate non-potato clades and are alternatively treated in sect. Etuberosum (Buk. and Kameraz) A. Child, sect. Lycopersicum (Mill.) Wettst., or sect. Juglandifolium (Rydb.) A. Child (Child, 1990
; Spooner, Anderson, and Jansen, 1993
; Contreras and Spooner, 1999
). There are several disagreements regarding the number of species and their assignment to series throughout sect. Petota (Spooner and Sytsma, 1992
; Spooner and van den Berg, 1992a
).
Our study examines species boundaries of all six tetraploid (2n = 4x = 48) wild potato species in Solanum ser. Longipedicellata Juz., all of which have Endosperm Balance Numbers (EBN, see below) of 2. The standard citation of this ploidy and EBN combination is 4x(2EBN). These six species are S. fendleri A. Gray, S. hjertingii Hawkes, S. matehualae Hjert. & T.R. Tarn, S. papita Rydb., S. polytrichon Rydb., and S. stoloniferum Schltdl. & Bouchet. They grow in the southeastern United States (S. fendleri) and Mexico (all six species) and constitute all the members of ser. Longipedicellata except for S. x vallis-mexici Juz., a localized triploid of putative hybrid origin between S. verrucosum Schltdl. (ser. Tuberosa Rydb. [Hawkes]) and S. stoloniferum (Hawkes, 1990
).
The treatment of section Petota by Hawkes (1990)
is the latest of many often-conflicting attempts at formal infrasectional classifications (Spooner and Sytsma, 1992
; Spooner and van den Berg, 1992a
). Hawkes's (1990)
series classification has been discordant with phylogenies from chloroplast DNA restriction site data (cpDNA; Spooner and Sytsma, 1992
; Spooner and Castillo, 1997
), and Amplified Fragment Length Polymorphisms (AFLPs; Kardolus, 1998
). The cpDNA data group all members of ser. Longipedicellata into a large clade that also includes all other Mexican and Central American polyploid species, the diploid Mexican species S. verrucosum, and most of the South American species. The AFLP data place all members of ser. Longipedicellata into a single clade, sister to S. verrucosum; sister to this clade is S. brachycarpum Correll (ser. Demissa); sister to this clade are South American species. Morphological data show S. fendleri (ser. Longipedicellata) to be distinguished from other South American species in the Solanum brevicaule complex only by character states that overlap in range (van den Berg et al., 1998
). In summary, members of ser. Longipedicellata are difficult to distinguish from members of other series in Mexico and in South America. It may be a natural group based partly on crossability data (below), shared tetraploid genomes, and restriction to North America and Mexico, but this is unresolved. We designed our study to explore the status of the six species in ser. Longipedicellata and the morphological distinction of this series from similar series.
We analyzed morphologically similar species in ser. Demissa Juz. [S. demissum Lindl. 6x(4EBN)] and ser. Tuberosa [S. avilesii Hawkes & Hjert., S. gourlayi Hawkes, and S. verrucosum, all 2x(2EBN)], and morphologically more dissimilar species in ser. Tuberosa [S. berthaultii, 2x(2EBN)] and ser. Yungasensia Correll [S. chacoense Bitter, 2x(2EBN)]. We chose S. demissum and S. verrucosum as Mexican representatives and S. avilesii and S. gourlayi as South American representatives of other series that are morphologically similar to members of ser. Longipedicellata. Solanum demissum was examined by Spooner, van den Berg, and Bamberg (1995)
with morphological data, and forms a close phenetic group with the South American ser. Acaulia, rather than with ser. Demissa. We re-examine it here because of morphological similarity to S. stoloniferum, S. fendleri, and S. verrucosum (see below).
Our study is designed to be compared to a similar morphological study of the Solanum brevicaule complex, a group of 
30 South American wild potato taxa, including diploids, tetraploids, and hexaploids, that are very similar to the cultivated potato. The species boundaries of this complex were investigated by van den Berg, Groendijk-Wilders, and Kardolus (1996)
, and van den Berg et al. (1998)
with morphological data, and Miller and Spooner (1999)
with molecular data. The two latter studies were concordant in suggesting that these 30 taxa could at best be partitioned into three species. These studies showed that although S. fendleri could be distinguished from members of this complex, this could only be done with multivariate techniques using quantitative character states that overlap in range. The present study includes four species as "placeholders" that were investigated in common between these studies: S. fendleri (ser. Longipedicellata), S. avilesii, S. gourlayi, and S. verrucosum (ser. Tuberosa).
Morphological criteria and species boundaries
Correll (1962)
and Hawkes (1990)
provided the latest comprehensive treatments of sect. Petota and therefore will form the basis for comparison to our study. They had slightly different concepts regarding affiliations of species to ser. Longipedicellata, ser. Demissa, and ser. Tuberosa. Correll (1962)
distinguished ser. Longipedicellata with "more or less petiolulate leaflets" and with "rotate-stellate corollas," and ser. Demissa (including S. verrucosum) with "corolla[s] mostly rotate-pentagonal, sometimes rotate-substellate or rotate-stellate, occasionally somewhat 10–15-lobulate." He keyed out members of ser. Tuberosa (Correll placed most species assigned here by Hawkes [1990
] in ser. Transaequatorialia Buk. ex Kameraz) largely by geographical data and provided no species-specific characters. Hawkes (1990)
distinguished ser. Longipedicellata by having a corolla with a "completely circular appearance with acumens standing out sharply from it," or with corollas "occasionally substellate to pentagonal." He stated that "the corolla lobes [of ser. Tuberosa] are not formed so as to give a circular appearance to the flower." Hawkes (1990)
distinguished ser. Demissa (excluding S. verrucosum placed in ser. Tuberosa by Correll [1962
]) by "corolla lobes very short and flat, with small acumens (1.5–2 mm long), giving the flowers a ten-lobed appearance."
Members of these series are morphologically similar to each other in floral and vegetative traits, and the species distinctions can be vague. Gray (1886)
noted the similarity of S. fendleri to the cultivated potato S. tuberosum L. (ser. Tuberosa) and synonymized it under S. tuberosum var. boreale A. Gray. Problems in distinguishing members of different series also were noted by Correll (1962
, p. 364): "In my opinion, the confusing of some plants of S. verrucosum with some plants of S. demissum is excusable. In fact, many plants collected in nature cannot be placed with certainty into either category, while specimens of cultivated plants of both S. verrucosum and S. demissum are exceedingly variable." Correll (1962
, pp. 380–382) continued: "Vegetatively, many plants of S. fendleri, S. demissum, S. verrucosum, and S. stoloniferum found in nature approach one another very closely, and because of this I have leaned heavily on the shape of the corolla for separation of these plants...It must be noted, however, that variations are frequent and must be taken into consideration...All of these species, however, are exceedingly variable, especially in the size of the corolla, and the amount of pubescence present on the plant."
Both authors also had slightly different concepts regarding species boundaries within ser. Longipedicellata. We summarize the characters they use in the keys and descriptions in Table 1 and provide a comparison of their species boundaries and affiliations to species and series in Fig. 1.
|
|
Genome hypotheses of ser. Longipedicellata
Within sect. Petota and the closest non-tuber-bearing relatives in section Etuberosum, chromosome pairing relationships have been interpreted by genome formulae (Marks, 1955
; Matsubayashi, 1955, 1991
; Hawkes, 1958, 1990
; Irikura, 1976
). Most authors agree on a five-genome hypothesis, although they do not use standard letter designations. The latest review of genome hypotheses by Matsubayashi (1991)
designates genomes as A, B, C, D, and E. The A genome and its superscripted minor variants are part of the genomes of all diploid and polyploid species except for the non-tuber-bearing species of Solanum section Etuberosum that are designated as E genome species. The B genome is associated with the Mexican tetraploid wild species of ser. Longipedicellata; the C genome is associated with the Mexican, Central, and South American wild species of the ser. Conicibaccata; and the D genome with the Mexican hexaploid wild species of ser. Demissa.
Meiotic studies show the tetraploid species of ser. Longipedicellata to form 24 bivalents at metaphase I and generally to form 12 bivalents and 12 univalents in artificial hybrids with a typical A genome diploid, S. chacoense. Matsubayashi (1955, 1991)
, Irikura (1976)
, Ramanna and Hermsen (1979)
, and Hawkes (1990)
postulated that members of ser. Longipedicellata are strict allotetraploids, with distinct A and B genomes. Hawkes (1990)
suggested that the B genome was a primitive indigenous genome from Mexico, that the A genome was a recent migrant from South America, and that the members of ser. Longipedicellata are allopolyploid derivatives of hybridization between these A and B genome diploids. See Spooner, van den Berg, and Bamberg (1995)
for a summary of these genome formulae by different authors for ser. Longipedicellata, ser. Demissa, and ser. Tuberosa.
Endosperm Balance Numbers
A major isolating mechanism in section Petota is governed by a strong crossing barrier evidenced by endosperm breakdown in hybrids differing by EBN compatibilities, referred to as the Endosperm Balance Number (EBN) hypothesis (Johnston et al., 1980
; Hanneman, 1994
). Species are assigned an EBN based on their ability to cross within EBN levels, using standard tester species. Within sect. Petota, species are 2x(1EBN), 2x(2EBN), 4x(2EBN), 4x(4EBN), and 6x(4EBN). Doubling ploidy doubles EBN (Hanneman, 1994
). The 4x(2EBN) condition is rare in sect. Petota, possessed by all members of ser. Longipedicellata, but only two other Mexican and Central American species, S. agrimonifolium Rydb. and S. oxycarpum Scheide in Schltdl. (members of ser. Conicibaccata), and a few other South American species.
Crossing relationships among species of ser. Longipedicellata
Hawkes (1966, 1990)
and Matsubayashi (1991)
documented that artificial interspecific hybrids within ser. Longipedicellata are fully interfertile. This contrasts sharply with general lack of intercrossability among species in the Mexican and Central American 6x(4EBN) species of ser. Demissa. In summary, species of ser. Longipedicellata are freely intercrossable with each other but have a strong EBN-based biological isolating mechanism from members of ser. Demissa and ser. Tuberosa and possess unique (putative) AABB genomes.
Purpose of the present study
This study contributes to our goal to produce a monograph of Solanum sect. Petota for North and Central America. It follows a morphological study of the six Mexican and Central American species in ser. Demissa (Spooner, van den Berg, and Bamberg, 1995
), chloroplast DNA (cpDNA) restriction site studies of the 30 wild potato species of North and Central America (Spooner and Sytsma, 1992
), cpDNA studies of the Mexican and Central American species S. bulbocastanum and S. cardiophyllum (Rodríguez and Spooner, 1997
), and field trips to Mexico and Central America (Spooner et al., 1991, 1998, 2000
, in press; Rodríguez et al., 1995
). It also is part of a goal to better define species boundaries and relationships of wild potatoes to better classify the holdings of the world's genebanks. We pose the following two questions in this study: (1) What is the pattern of character state variation in the six tetraploid species of ser. Longipedicellata, and does this provide morphological support for these species? (2) Is there a practical taxonomic separation of members of ser. Longipedicellata from phenetically similar members of ser. Tuberosa using S. avilesii, S. gourlayi, and S. verrucosum as placeholders in reference to an earlier study of the S. brevicaule complex (van den Berg et al., 1998
; Miller and Spooner, 1999
), and to S. demissum (ser. Demissa)?
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) and because of the uncertainty of many of these identifications. We also analyzed morphologically similar species in ser. Demissa (S. demissum, 4), ser. Tuberosa (S. avilesii, 2; S. gourlayi, 3; and S. verrucosum, 3); and yet morphologically more dissimilar species in ser. Tuberosa (S. berthaultii, 1) and ser. Yungasensia (S. chacoense, 1).
|
|
). They represent the maximum geographic distribution available from the NRSP-6 collection (Fig. 2), and nearly the entire geographic ranges for these species as mapped by Hawkes (1966)
. Vouchers are deposited at the NRSP-6 herbarium in Sturgeon Bay, Wisconsin (PTIS, see http://www.nybg.org/bsci/ih/ih.html). Identifications of these accessions have been provided in past years by visiting taxonomists during on-site visits to NRSP-6 to identify living representatives in field plots. To aid in communicating intra-accession variation and interspecific similarity of all six species within ser. Longipedicellata and morphologically similar species in ser. Tuberosa (S. avilesii, S. gourlayi, and S. verrucosum), and ser. Demissa (S. demissum) we show leaf variation (Fig. 3). We selected these accessions to illustrate variation within and between accessions, species, and series.
|
Data measurement
The middle three of the six plants per row were measured for each accession. We measured 47 quantitative and seven qualitative characters (Table 3) when the plants were in full bloom in August, or in September for fruits, that set naturally in the field. Trichome density (characters 3–4) was assessed from young, fully expanded leaves in the interveinal areas, using a hand-held clear acetate strip with different numbers of points per square centimetre. Leaf measurements were made on the largest leaf per plant. Floral characters were measured on the uppermost inflorescence. Corolla colors were measured with the aid of the Royal Horticultural Society Colour Charts (Royal Horticultural Society, 1986
), based on recommendations of Tucker, Maciarello, and Tucker (1991)
. Because of varying colors and color intensities from blue to violet and white, and difficulty of ordering the colors by eye, we determined color intensity from charts with a Minolta chroma meter CR-221TM. The 48 scored RHS colors, ordered by their corresponding intensity values, are: 155A (white, 102.3), 91D (94.1), 92D (93.1), 85D (93.0), 84D (92.3), 91C (86.2), 84C (83.6), 92C (82.7), 85C (82.2), 91B (77.7), 81A (76.0), 85B (75.2), 84B (74.3), 95D (72.1), 92B (71.9), 85A (69.6), 93D (66.8), 91A (63.4), 94C (63.1), 86D (59.9), 90D (59.0), 84A (57.3), 88C (56.6), 92A (55.7), 90C (53.3), 82A (52.1), 94B (50.8), 83D (50.4), 86C (50.2), 93C (49.1), 88B (47.5), 90B (45.2), 86B (44.8), 88A (43.2), 83C (42.4), 98C (41.0), 77A (39.0), 90A (38.3), 93B (37.3), 94A (37.2), 83B (33.9), 86A (31.4), 79B (31.0), 83A (27.9), 79A (25.1), 93A (24.6), 89C (23.2), 89B (dark purple, 17.9). Raw data are available from the authors upon request.
|
). Dendrograms including all accessions were produced by NTSYS-pcR version 1.70 (Rohlf, 1992
). Averages for each character were standardized (STAND) and similarity matrices (in SIMINT), using average taxonomic distance (DIST), Manhattan distance (MANHAT), and product-moment correlation (CORR) were generated. Clustering was performed using the unweighted pair-group method (UPGMA) in SAHN. Cophenetic correlation coefficients (COPH and MXCOMP) were used to measure distortion between the similarity matrices and the resultant three phenograms (Rohlf and Sokal, 1981
; Sokal, 1986
). Principal components analysis (PCA), canonical discriminant analysis (CDA), and stepwise discriminant analysis (SDA) were performed by SAS Version 7 (SAS, 1998
) using PRINCOMP, CANDISC, and STEPDISC, with the means of six individuals per accession. The PCA, CDA, and SDA were run twice, once with all taxa and all characters. They were run again with only the six species of ser. Longipedicellata and with a reduced character set that showed significant differences among any two pairs of species in ser. Longipedicellata, using the Tukey-Kramer HSD test in JMP software (SAS, 1995
). The dendrograms were produced only with all taxa and all characters.
PCA and CDA are both ordination techniques, but PCA makes no assumptions about group membership of OTUs. It attempts to portray multidimensional variation in the data set in the fewest possible dimensions, while maximizing the variation. CDA uses assigned groups to derive a linear combination of the variables (morphological characters) that produces the greatest separation of the groups (SAS, 1998
). Cluster analysis, like PCA, makes no assumptions about group membership; it produces trees based on average similarity of all data. It is convenient graphically for studies such as this one with many OTUs, because of ease of labeling the terminal branches. PCA and dendrograms therefore are more appropriate to explore phenetic structure without any assumptions of species boundaries, while CDA is an appropriate technique to test preexisting classifications. Because these phenetic approaches use very different algorithms and operate under very different assumptions about the data set, we used all three analyses in our exploration of phenetic structure in Solanum ser. Longipedicellata and the other species outside of this series.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
stated that cophenetic correlations between 0.8 and 0.9 could be interpreted subjectively as good fits to the cluster analysis, and those between 0.7 and 0.8 as poor fits. It clearly separates the morphologically most dissimilar species, S. chacoense and S. berthaultii, at the very base of the tree. All four accessions of S. demissum cluster near the base of the tree. The other non-ser. Longipedicellata species cluster among ser. Longipedicellata species. All three accessions of S. verrucosum cluster. The two accessions of S. avilesii and three accessions of S. gourlayi (ser. Tuberosa) are intercalated in the members of ser. Longipedicellata. Regarding ser. Longipedicellata, S. matehualae and S. hjertingii form a cluster. All other species, S. fendleri, S. papita, S. polytrichon, and S. stoloniferum are intermixed on the tree.
|
A PCA of the reduced data set (only ser. Longipedicellata, only 38 of 54 characters) is presented in Fig. 7. Principal components 1, 2, and 3 account for 20, 13.4, and 10.4% of the variation, respectively, for a total of 43.8%. PCA axis 1 is most highly influenced by (presented in decreasing order of loadings) (1) radius of corolla, (2) length of stamen, (3) pedicel length, (4) length of style, (5) number of interjected leaflets; PCA axis 2 by (1) ratio: length of terminal leaflet lamina from its widest point to apex/length of lamina, (2) length of terminal leaflet lamina, (3) pedicel pubescence posture, (4) length of calyx acumen, (5) width of most distal lateral leaflet 5 mm below apex; PCA axis 3 by (1) length of style, (2) length of style exsertion from apex of anther column, (3) ratio: length of terminal leaflet lamina/width of lamina, (4) density of adaxial pubescence, (5) width of terminal leaflet 5 mm below apex. It somewhat separates (1) S. polytrichon, (2) S. hjertingii + S. matehualae, (3) S. fendleri + S. papita + S. stoloniferum.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
It is difficult to make firm generalizations about the characters most affecting structure of the ser. Longipedicellata PCA and CDA because of mixture of character types on different axes. However PCA axis 1 generally is most highly influenced by size of flower parts, PCA axis 2 by terminal leaflet shape and size, and PCA axis 3 by style length. The CDA axis 1 is most highly influenced by style length and curvature, and CDA axis 2 by pubescence of pedicel and leaf. The results of the two most important characters of the SDA are illuminating. The first character, length of style, characterizes S. hjertingii + S. matehualae, while the second character, pedicel pubescence, characterizes S. polytrichon.
Species distinctions are further complicated by intra-accession variation, illustrated in the leaves of S. hjertingii, S. papita, and S. polytrichon (ser. Longipedicellata), S. avilesii, S. gourlayi, and S. verrucosum (ser. Tuberosa), and S. demissum (ser. Demissa) (Fig. 3). Correll (1962
, his plates VIII–X) dramatically illustrated such intra-accession variation in three specimens of S. demissum (Correll 14238), one collected in the field in Mexico, one grown from tubers of this accession in a greenhouse in Maryland, and another grown from tubers of the same accession in a field in Wisconsin. The taxonomic difficulty of sect. Petota comes into focus considering such intra-accession variation and the lack of clear diagnostic characters separating species.
Phenetic structure of ser. Longipedicellata relies entirely on polythetic support, i.e., grouping taxa that have the greatest number of shared features, no single feature of which is essential for group membership or is sufficient to make an organism a member of a group (Sokal and Sneath, 1963
; Stuessy, 1990
). This pattern is proving to be the rule in species in sect. Petota (Spooner and van den Berg, 1992b
; Giannattasio and Spooner, 1994a, b
; Spooner, van den Berg, and Bamberg, 1995
; Castillo and Spooner, 1997
; van den Berg et al., 1998
). Clearly, keys and descriptions of species that ultimately are recognized in ser. Longipedicellata will need to account for the intra-accession and interspecific variation inherent in the group.
Our results make a clear case for the reduction of the six species of ser. Longipedicellata to the three species S. hjertingii, S. polytrichon, and S. stoloniferum (the earliest names). We suspect that these will be the species we ultimately recognize in our potato flora of North and Central America, but we await a formal taxonomic treatment of these species for four reasons. First, our morphological data are from germplasm accessions grown in an atypical environment. However, this allowed us to measure all organs, including delicate floral organs that often are missing or poorly preserved on herbarium specimens and leaves of comparable size, and our results match our intuitive impressions of herbarium specimens collected from the wild. Second, we wish to be conservative in making taxonomic changes in this economically important group as the scientific literature reports on evaluations of wild potato species literally every month. Third, we await results of our molecular studies on this group. However, cpDNA studies of all of these species (Spooner and Sytsma, 1992
; Spooner and Castillo, 1997
) groups them in polytomies or includes them all in the same clade and separates them only on small branches. Fourth, we wish to study type material to insure that our collections are properly identified.
Distinction of ser. Longipedicellata to ser. Demissa and ser. Tuberosa
Our measurements of the corolla shape and petiolule characters used by Correll (1962)
and Hawkes (1990)
for ser. Longipedicellata, ser. Demissa and ser. Tuberosa fail to distinguish them, and our data show no series-specific character states (Fig. 4). Morphology, therefore, does not provide any absolute character states to clearly separate members of ser. Longipedicellata from our placeholder members of ser. Tuberosa (S. avilesii, S. gourlayi, and S. verrucosum), or ser. Demissa (S. demissum). The morphological study of the S. brevicaule complex (van den Berg et al., 1998
) similarly showed the placeholder taxa of ser. Longipedicellata and ser. Tuberosa examined here to be distinguished from each other only by character states with overlapping ranges, not by more practical breaks between quantitative character states or by qualitative characters. This is illustrated in representative leaves of members of all of these series (Fig. 3) that show great similarity of overall form and no evident character states separating series, a pattern typical of all other character states in these series. There are simply few if any qualitative characters useful for distinguishing these species and even series. Our present results help to document and explain the reason for the continuing disagreement among taxonomists for species and series boundaries in the group (Spooner and van den Berg, 1992a
).
Our results provide no answer to the question of the evolution of members of ser. Longipedicellata. The AABB genome hypothesis suggests an allopolyploid origin of the series, but the parental species are unknown. Even if they were known, McDade (1990)
and Rieseberg and Ellstrand (1993)
point out that hybridity is not always associated with additive morphology. Our results show extensive similarity of species across three currently recognized series, suggesting that it would be difficult to find clear species-specific morphological markers for putative progenitors of the series. We can only speculate about origins based on other data. Ortiz and Ehlenfeldt (1992)
use EBN to speculate on the origin of the series by bilateral sexual polyploidization of two 2x(1EBN) species, or alternatively by bilateral sexual polyploidization of two 2x(2EBN) species followed by a reduction in EBN. None of these hypotheses, however, has ever been tested. The allopolyploid origin of the series can also be tested at the molecular level by searching for additivity of species-specific molecular markers (e.g., Dvorak and Zhang, 1992
), or at the cytological level by genomic in-situ hybridization, GISH (e.g., Bennett, Kenton, and Bennett, 1992
; Jiang and Gill, 1994
).
Despite the great morphological similarity of members of ser. Longipedicellata to some members of ser. Tuberosa, they are reproductively isolated based on their 4x(2EBN) crossability, and possibly by their AABB genome constitution. This supports S. hjertingii, S. polytrichon, and S. stoloniferum as worthy of taxonomic recognition separate from the morphologically similar species in Mexico and South America. This example clearly demonstrates a conflict of a morphological and a biological species concept, and the continuing difficulty in the use of morphology for the preparation of keys and descriptions, and identification of specimens. Earlier we (Spooner and van den Berg, 1992a
) criticized the use of geographical characters in keys of Correll (1962)
and Hawkes (1990)
to distinguish species. However, considering the difficulty of morphology to distinguish good biological species, this may be the only practical solution in key construction. A practical taxonomic treatment of the group using morphology likely will always be an elusive goal.
|
|
|
|
|
FOOTNOTES |
|---|
5 Author for reprint requests (dspooner{at}facstaff.wisc.edu
).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Bennett, S. T., A. Y. Kenton, and M. D. Bennett. 1992 Genomic in situ hybridization reveals the allopolyploid nature of Milium montianum (Gramineae). Chromosoma 101: 420–424[CrossRef][ISI]
Castillo, R. O., and D. M. Spooner. 1997 Phylogenetic relationships of wild potatoes, Solanum Series Conicibaccata (Sect. Petota). Systematic Botany 22: 45–83
Child, A. 1990 A synopsis of Solanum subgenus Potatoe (G. Don) (D'Arcy) (Tuberarium (Dun.) Bitter (s.l.)). Feddes Repertorium 101: 209–235
Contreras M., and D. M. Spooner. 1999 Revision of Solanum section Etuberosum (subgenus Potatoe). In M. Nee, D. E. Symon, R. N. Lester, and J. P. Jessop [eds.], Solanaceae IV, Advances in Biology and Utilization, 227–245. Royal Botanic Gardens, Kew, UK
Correll, D. S. 1962 The potato and its wild relatives. Contributions from the Texas Research Foundation, Botanical Studies 4: 1–606
Dvorak, J., and H. B. Zhang. 1992 Application of molecular tools for study of the phylogeny of diploid and polyploid taxa in Triticeae. Hereditas 116: 37–42[CrossRef][ISI]
Giannattasio, R. B., and D. M. Spooner. 1994a A reexamination of species boundaries between Solanum megistacrolobum and S. toralapanum (Solanum sect. Petota, series Megistacroloba): morphological data. Systematic Botany 19: 89–105
———, and ———. 1994b A reexamination of species boundaries and hypotheses of hybridization concerning Solanum megistacrolobum and S. toralapanum (Solanum sect. Petota, series Megistacroloba): molecular data. Systematic Botany 19: 106–115
Gray, A. 1886 Synoptical flora of North America, vol. 2, part 1. Ivison, Blakeman, Taylor and Company, London, UK
Hanneman, R. E., Jr. 1994 Assignment of Endosperm Balance Numbers to the tuber-bearing Solanums and their close non-tuber-bearing relatives. Euphytica 74: 19–25[CrossRef][ISI]
Hawkes, J. G. 1958 Kartoffel. I. Taxonomy, cytology and crossability. In H. Kappert and W. Rudorf [eds.], Handbuch Pflanzenzüchtung, 2nd ed., 1–43. Paul Parey, Berlin, Germany
———. 1966 Modern taxonomic work on the Solanum species of Mexico and adjacent countries. American Potato Journal 43: 81–103
———. 1990 The potato: evolution, biodiversity and genetic resources. Belhaven Press, London, UK
Irikura, Y. 1976 Cytogenetic studies on the haploid plants of the tuber-bearing Solanum species. II. Cytogenetical investigations on haploid plants and interspecific hybrids by utilizing haploidy (in Japanese, with English summary). Research Bulletin of the Hokkaido National Agricultural Research Station 115: 1–80
Jiang, J., and B. S. Gill. 1994 Different species-specific chromosome translocations in Triticum timopheevii and T. turgidum support diphyletic origin of polyploid wheats. Chromosome Research 2: 59–64[CrossRef][Medline]
Johnston, S. A., T. P. M. Den Nijs, S. J. Peloquin, and R. E. Hanneman, Jr. 1980 The significance of genetic balance to endosperm development in interspecific crosses. Theoretical and Applied Genetics 53: 262–269
Kardolus, J. P. 1998 A biosystematic analysis of the polyploid series Acaulia and related potato species. Chapter 5 of A biosystematic analysis of Solanum acaule, 63–84. Ph.D. dissertation, Wageningen Agricultural University, Wageningen, The Netherlands
Marks, G. E. 1955 Cytogenetic studies in tuberous Solanum species. I. Genomic differentiation in the group Demissa. Journal of Genetics 53: 262–269
Matsubayashi, M. 1955 Studies on the species differentiation of the section Tuberarium of Solanum III. Behavior of meiotic chromosomes in F1 hybrid between S. longipedicellatum and S. schickii in relation to its parent species. Science Reports of the Hyogo University of Agriculture 2: 25–31
———. 1991 Phylogenetic relationships in the potato and its related species. In T. Tsuchiya and P. K. Gupta [eds.], Chromosome engineering in plants: genetics, breeding, evolution, part B: 93–118. Elsevier, Amsterdam, The Netherlands
McDade, L. 1990 Hybrids and phylogenetic systematics I. Patterns of character expression in hybrids and their implications for cladistic analysis. Evolution 44: 1685–1700[CrossRef][ISI]
Miller, J. T., and D. M. Spooner. 1999 Collapse of species boundaries in the wild potato Solanum brevicaule complex (Solanaceae, S. sect. Petota): molecular data. Plant Systematics and Evolution 214: 103–130[CrossRef][ISI]
Ortiz, R., and M. K. Ehlenfeldt. 1992 The importance of Endosperm Balance Number in potato breeding and the evolution of tuber-bearing Solanum species. Euphytica 60: 105–113[ISI]
Ramanna, M. S., and J. G. T. Hermsen. 1979 Genome relationships in tuber-bearing Solanums. In J. G. Hawkes, R. N. Lester, and A. D. Skelding [eds.], The biology and taxonomy of the Solanaceae, 647–654. Linnean Society of London Symposium Series 7. Academic Press, London, UK
Rieseberg, L. H., and N. C. Ellstrand. 1993 What can molecular and morphological markers tell us about plant hybridization? Critical Reviews in Plant Sciences 12: 213–241[CrossRef]
Rodriguez, A., and D. M. Spooner. 1997 Chloroplast DNA analysis of Solanum bulbocastanum and S. cardiophyllum, and evidence for the distinctiveness of S. cardiophyllum subsp. eherenbergii (sect. Petota). Systematic Botany 22: 31–43
———, O. Vargas, E. Villegas, and D. M. Spooner. 1995 Wild potato (Solanum sect. Petota) germplasm collection expedition to Mexico in 1993, with special reference to Solanum bulbocastanum Dunal and S. cardiophyllum Lindley. Potato Research 38: 47–52[CrossRef][ISI]
Rohlf, F. J. 1992 NTSYS-pc, numerical taxonomy and multivariate system. Exeter Publishing, Ltd., New York, New York, USA
———, and R. R. Sokal. 1981 Comparing numerical taxonomic studies. Systematic Zoology 30: 459–490[CrossRef]
Royal Horticultural Society. 1986 R.H.S. colour chart. [ed. 1, 2]. Royal Horticultural Society, London, UK
SAS. 1995 JMP Software, Version 3.1. SAS Institute Inc., Cary, North Carolina, USA
———. 1998 SAS Proprietary Software, Version 7. SAS Institute Inc., Cary, North Carolina, USA
Sokal, R. R. 1986 Phenetic taxonomy: theory and methods. Annual Review of Ecology and Systematics 17: 423–442[CrossRef][ISI]
———, and R. R. Sneath. 1963 Principles of numerical taxonomy. Freeman, San Francisco, California, USA
Spooner, D. M., G. J. Anderson, and R. K. Jansen. 1993 Chloroplast DNA evidence for the interrelationships of tomatoes, potatoes, and pepinos (Solanaceae). American Journal of Botany 80: 676–688[CrossRef][ISI]
———, J. Bamberg, J. P. Hjerting, and J. Gomez. 1991 Mexico, 1988 potato germplasm collecting expedition and utility of the Mexican potato species. American Potato Journal 68: 29–43[ISI]
———, and R. Castillo T. 1997 Reexamination of series relationships of South American wild potatoes (Solanaceae: Solanum sect. Petota): evidence from chloroplast DNA restriction site variation. American Journal of Botany 84: 671–685[Abstract]
———, R. Hoekstra, R. G. Van Den Berg, and V. Martinez. 1998 Solanum sect. Petota in Guatemala; taxonomy and genetic resources. American Journal of Potato Research 75: 3–17
———, ———, and B. Vilchez. In press. Solanum sect. Petota in Costa Rica: taxonomy and genetic resources. American Journal of Potato Research.
———, A. Rivera-Pena, R. G. Van Den Berg, and K. Schuler. 2000 Potato Germplasm Collecting Expedition To Mexico In 1997: taxonomy and new germplasm resources. American Journal of Potato Research 77: 261–270[ISI]
———, and K. J. Sytsma. 1992 Reexamination of series relationships of Mexican and Central American wild potatoes (Solanum sect. Petota): evidence from chloroplast DNA restriction site variation. Systematic Botany 17: 432–448[CrossRef][ISI]
———, and R. G. Van Den Berg. 1992a An analysis of recent taxonomic concepts in wild potatoes (Solanum sect. Petota). Genetic Resources and Crop Evolution 39: 23–37[CrossRef]
———, and ———. 1992b Species limits and hypotheses of hybridization of Solanum berthaultii Hawkes and S. tarijense Hawkes: morphological data. Taxon 41: 685–700[CrossRef][ISI]
———, ———, and J. Bamberg. 1995 Examination of species boundaries of Solanum series Demissa and potentially related species in series Acaulia and series Tuberosa (sect. Petota). Systematic Botany 20: 295–314[CrossRef][ISI]
Stuessy, T. F. 1990 Plant taxonomy: the systematic evaluation of comparative data. Columbia University Press, New York, New York, USA
Tucker, A. O., M. J. Maciarello, and S. S. Tucker. 1991 A survey of color charts for biological descriptions. Taxon 40: 201–214[CrossRef][ISI]
Van Den Berg, R. G., N. Groendijk-Wilders, and J. P. Kardolus. 1996 The wild ancestors of cultivated potato: the brevicaule-complex. Acta Botanica Neerlandica 45: 157–171
———, J. T. Miller, M. L. Ugarte, J. P. Kardolus, J. Villand, J. Nienhuis, and D. M. Spooner. 1998 Collapse of morphological species in the wild potato Solanum brevicaule complex (Solanaceae: sect. Petota). American Journal of Botany 85: 92–109[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
= S. hjertingii, 
= S. matehualae; 
= S. polytrichon; * = S. papita; 
= S. stoloniferum; v = S. verrucosum.
