Author: Comish, P. S.

Author affiliation: Hortic. Res. Advis. Stn., NSW Agric., Gosford, Australia.

Published in: Australian Journal of Experimental Agriculture, volumn 32, pages 395-399, (1992).

Abstract: "Glyphosate residues in a loamy sand soil were suspected of damaging transplanted tomatoes at Gosford in 1990. Field and glasshouse expts. were conducted to det. whether phytotoxic residues of glyphosate persist in this soil type and, if so, under what conditions. In the glasshouse expt, visible symptoms of glyphosate toxicity occurred in tomato seedlings transplanted into soil that was sprayed 1, 5 or 15 days earlier with glyphosate (360 g/L) at 4 L product/ha. Glyphosate also reduced plant dry wt. (16 days after transplanting), but only where soil nutrient deficiencies were car. after transplanting. In this case,
seedlings transplanted 15 days after spraying suffered an ay. redn. in dry wt. of 57%. Greater redns. in dry wt. occurred where superphosphate (43 kg P/ha) was mixed through soil before spraying (75 v. 35% redn.). In the field, glyphosate residues reduced plant dry wt. 16 days after transplanting, even when transplanting followed spraying by up to 9 days, and possibly as many as 30. At 9 days, redns. of 50, 74 and 78% were recorded with glyphosate (360 g/L) applied at 2, 4 and 8 L/ha, resp. Effects of glyphosate on fruit yield were significant, but much smaller than effects on earlier plant dry wts. The phytotoxicity of glyphosate residues in this loamy sand results from a combination of inherently low P sorption capacity and application of superphosphate, leading to low adsorption of glyphosate by soil. This may be exacerbated when dry conditions occur between application and planting. Thus, a plant-back period of 3 wk could be considered safe when transplanting tomatoes into this sandy soil, provided some mixing of soil occurs at transplanting. It is recommended that farmers perform a simple bioassay to confirm safety."
--------------------------------

Title: Initial and residual activity of glyphosate and SC-0224 in a sandy soil.

Authors: Devlin, Robert M.; Karczmarczyk, Stanislaw J.; Zbiec, Irena I.; Koszanski, Zdzislaw K.

Authors affiliation: Lab. Exp. Biol., Univ. Massachusetts, East Wareham, MA, USA.

Published in: Crop Protection, volumn 5, pages 293-296, (1986).

Abstract: "Initial and residual herbicide activity of glyphosate [1071-83-6] and SC-0224 (trimethylsulfonium carboxymethylaminomethylphosphonate) [81591-81-3] when applied to a sandy soil was investigated. A bioassay employing wheat (Triticum vulgare cv Mericopa) was used to det. the residual activity of the herbicides on different characteristics of plant growth. At 5 kg/ha, both herbicides significantly reduced shoot length. This was obsd. in wheat planted immediately as well as 10 days after application. SC-0224 was more active than glyphosate, significantly reducing shoot length at the 2.5 kg/ha rate 10 days after application. Root length was reduced by both herbicides at 2.5 and 5 kg/ha in wheat planted 10 days after application. Redns. in fresh and dry wt. were also obsd. for both shoots and roots. Thus, SC-0224 and glyphosate have residual herbicide activity in sandy soil, at least �10 days after application, and SC-0224 has more herbicidal activity than glyphosate in this respect."
----------------------
Title: Herbicidal activity of glyphosate in soil.

Authors: Salazar, Luis C.; Appleby, Arnold P.

Authors affiliation: Dep. Crop Sci., Oregon State Univ., Corvallis, OR, USA.

Published in: Weed Science, volumn 30, pages 463-466, (1982).

Abstract: "Greenhouse studies were conducted using bentgrass (Agrostis tenuis Highland) as a bioassay species to measure herbicidal activity of 3.4 kg/ha of the glyphosate isopropylamine salt [70768-30-8] when applied directly to the soil surface prior to bentgrass emergence. Activity of glyphosate also was examd. by applying glyphosate to a moist soil surface and placing alfalfa (Medicago sativa) and red clover (Trifolium pratense) seeds on the sprayed surface 3, 6, 9, and 24 h later. In 2 high-org. soils, glyphosate reduced bentgrass growth in each of four expts., even when applied to 5 days before bentgrass emergence. Significant glyphosate activity was measured in one of the expts. in another org. soil and in 3 of the expts. in a Chehalis sandy loam soil. Germination and growth of alfalfa and red clover were reduced when seeds were distributed on a sprayed soil surface �24 h after glyphosate application."

-------------------------------------------------

It can leave the roots of a sprayed plant, travel through the soil, and be picked up by the roots of nearby plants.

A 2006 key paper concerning transfer of Round-Up from leaves to roots, into the soil, and then being picked up by the roots of another plant is:
http://www.jpdp-online.com/Artikel.dll/02-Roemheld_MTAyNzEw.PDF
That paper was cited by 20 more recent scientific papers which include a 2007 one:

http://www.springerlink.com/content/q0941n83w14wq825/

This is part of the conclusion section of the 2007 paper:

"Second, due to translocation glyphosate can end up
to deeper layers or root zone quite rapidly compared
to leaching. The environmental impact of translocation
depends on the depth of the root zone that is
determined by the dominant weed species and the
hydraulic conditions of the field site. Rapid release
via roots into the rhizosphere or root channels may
result in fairly high concentrations of glyphosate. This
can cause various negative effects in non-glyphosate
resistant plants through root uptake (Rodrigues et al.
1982; Guldner et al. 2005; Neumann et al. 2006)."