Genetic Pollution - Asgrow Review

28 September 1992
Dr. Jane Rissler
Biotechnology Policy Center
National Wildlife Federation
1400 16th St. NW
Washington, DC 20036-2266
FAX: 202-797-6646

Dear Dr. Rissler:

This is in reference to your letter of 18 Sep 92 and the Asgrow petition. It is unfortunate that the primary focus of federal regulation in this area appears to be associated with weeds and pests within U.S. boundaries. The potential economic impact resulting from essentially unknown biological consequences of gene flow from transgfenic plants to wild relatives and landrace cultivars in areas of crop origin or diversity seem, in my view, to be much more significant than local production of weeds. At any rate, this review is mostly limited to areas that, from my limited knowledge of USDA regulations, appear to be significant in terms of current policy.

I feel that the petition is not fully consistent with published information regarding Cucurbita pepo in two areas: 1) identification of wild or free-living types, and 2) crop/weed gene flow. The nature of this inconsistency suggests, at least to me, that the bias is intended.

Identification of wild or free-living types:

Relatively recent work with the C. pepo 'complex' has produced a classification (Decker, 1988 - paper cited, but classification not mentioned in the petition) that reflects two fundamental elements:

1. C. pepo subsp. pepo - all domesticated types (Zucchini, Pumpkin, and others)

2. C. pepo subsp. ovifera - some domesticates and some wild types

Two taxonomic varieties, both involved in this petition, occur within subsp. ovifera:

1. subsp. ovifera var. ovifera - all domesticated (Crookneck [the topic of this petition], Acorn, 'Ornamental gourds', etc.)

2. subsp. ovifera var. texana - "all populations sustaining themselves in the natural environment. Current distribution includes Texas, Arkansas, Missouri, Alabama, and Illinois" (from Decker, 1988).

This classification reflects a perspective that places the Asgrow 'Crookneck' cultivar is part of one of the few domesticated plant lineages (subsp. ovifera var. ovifera) that have a center of origin in what is now the United States. This notion is well established by Decker (1988), developed in more detail by Decker-Walters (1990 and Decker-Walters et al, 1990 - not cited in the petition), and overviewed with distribution maps by Nee (1990 - not cited in the petition). This taxonomic perspective removes the associated wild type from the status of a species (C. texana) endemic to Texas. The wild element (subsp. ovifera var. texana) is now considered to be conspecific with all cultivars of C. pepo and consubspecific with a group of cultivars that includes the 'Crookneck' types. As indicated in the enclosed papers, this taxonomic alignment produces a picture of crop/weed relationships the U.S. that is quite different than that depicted in the petition.

The petition also ignores C. fraterna, a wild Mexican species that - as indicated in the enclosed papers - appears to have close connections to the Mexican lineage (C. pepo subsp. pepo).

Crop/weed gene flow crop/weed gene flow

The petition relies on data taken from Cucumis (Handel, 1982-83) for assessment of gene flow. Work with a specific focus on C. pepo (Kirkpatrick and Wilson, 1988;Wilson, 1990) is ignored. The study not cited documents a relatively high rate of crop/weed gene flow (ca. 5%) between populations that are separated by ca. 1300 meters. It also points out the significance of two native bee genera, which have evidently co-evolved with C. pepo (see enclosed papers).

With regard to specific questions:

i.) Weedy Squash? - If 'squash' means domesticated types, then YES the arguments presented in the petition are sound. 'Crookneck' types are highly evolved domesticates that do not carry a genetic background that would support 'escape' from cultivation or any type of competitive advantage in the wild. If genetic manipulations are as indicated in the petition, the transgenic line should be no different.

ii) Gene flow? - NO

1. The field study (fig. 2) lacks a population structure (few weeds at the margins or beyond the cultivated plot) that would be expected in a weed/crop situation.

2. The petition (fig. 2) does not indicate the number of fruits or progeny examined at each distance. The percentage values provided in fig. 2 could be misleading.

3. The field study does not indicate which pollinators were active in experimental plots. The presence or absence of 'squash' bees is significant.

iii) adverse effects? - NO

1. Viral resistance in the 'gene pool': The argument that wild types carry resistance is reasonable. However, I believe (I am not sure of this and do not have time to confirm) that the screens cited (Provvidenti, 1986, or possibly virus resistance trials published elsewhere) included 'C. texana' and the results indicated that 'C. texana' carries no resistance. Thus, while transfer would "..not represent an unique addition to the gene pool of wild Cucurbits..", it would be something new for the wild form of C. pepo.

2. Weedy tendencies and distribution of the wild types: As indicated by work cited in the petition (Harrison, et al. 1977), free-living C. pepo can become a weed problem. As indicated above, this free-living type shows a broad distribution in the eastern U.S. and, if references in the petition indicating 'ovifera' types in Georgia and Florida refer to wild or free-living types, then it might occur in these areas. Also, I believe that considerable C. pepo commercial production occurs in southern Texas (within the range of 'typical' C. pepo subsp. ovifera var. texana , but this is not indicated in the petition

iv) Scientific Standards: As indicated above, I feel that this petition does not present an objective picture of crop/weed gene flow in the Cucurbita pepo complex. Work not cited (listed below) can be extracted from just about any scientific database with key words 'Cucurbita" and "Gene flow". The most critical elements - taxonomic circumscription of the crop and its wild relatives and hybridization potential - are present in a paper cited in the petition (Decker, 1988). Subsequent elaboration (other papers cited below) has been published in journals that have a fairly high profile. Thus, the discussion seems to be biased, the bias seems to be self-serving and, as a result, the petition is not consistent with standards of thorough, objective literature review and presentation of data. It is a poor model.

I am sending a FAX copy of this and mailing the original with enclosed reprints. I hope that this will be of some help. I have no problem with your mention of my name as a reviewer of this petition.

Sincerely,

Hugh D. Wilson
Professor

Relevant papers not cited in the petition:

Decker-Walters, DS (1990) Evidence for multiple domestication of Cucurbita pepo. In DA Bates, RW Robinson, C Jeffrey [eds.], Biology and utilization of the Cucurbitaceae, pp. 96-101. Cornell University Press, Ithaca.

Decker-Walters, DS, TW Walters and U Posluszny (1990) Genealogy and gene flow among annual domesticated species of Cucurbita. Canadian Journal of Botany 68:782-789.

Kirkpatrick, K. J., and H. D. Wilson. 1988. Interspecific gene flow in Cucurbita: C. texana vs. C. pepo. Amer. J. Bot. 75:519-527.

Nee, M (1990) The domestication of Cucurbita (Cucurbitaceae). Econ. Bot. 44(3 supplement):56-68.

Wilson, HD (1990) Gene flow in squash species. Bioscience 40:449-455.

Return to Prologue		Move to the table of contents