# Dr. W. H. Summerson (Director, FDA Bureau of Science) — his National Conference paper attacking the Frawley proposal

**Provenance / caveat.** Summerson delivered this paper at FDA's **National Conference on Indirect Food
Additives, 13–14 Feb 1968** (where his paper "dealt almost solely with the Frawley proposal"). **The full
standalone paper is NOT among the 251 ToxicDocs "Frawley" documents** — those contain only the *Food
Chemical News* report that quotes it **"in part."** The fullest text available locally is reproduced below,
verbatim from that report: *Food Chemical News*, **19 Feb 1968, pp. 5–7** (reprinted to the SPI committee
by Keller & Heckman), local file `toxicdocs/1yes__1968__AlliedSignal__PolyvinylC__Kza60dbJ…`, Bates
**ASI-PR 0000576–0000578**. (Light `[sic]`/OCR notes added; the full paper would be in the FDA conference
proceedings — an unlocated, operator-retrieval item, see `05_OPEN_QUESTIONS`.)

---

## Headline (verbatim): *"Frawley Proposal Praised for 'Soundness,' Hit as 'Sheer Nonsense'"*

> Despite Summerson's concession that there is "soundness" in part of Frawley's proposal, the FDA-er said
> much of what Frawley proposed for a "single cut-off place" is "sheer nonsense."

## Summerson's paper "said, in part" (verbatim excerpt)

**(1) The cancer-latency argument for premarket testing:**
> "Unless premarketing clearance is practiced, the only method of detecting harmful effects is
> retrospective… The latent or 'incubation' period from the [time] a chemical agent is first applied to
> the human being until the time that cancer occurs is often 10 years or more… it is well nigh impossible
> to isolate an additive as the original causative agent of a cancer. Thus… we must require premarket
> testing."

**(2) A direct rebuke of the industry's "no known harm" posture:**
> "…even today we have offered to us a statement to the effect that the proponent feels no food additive
> clearance is necessary because he has no knowledge of any bad effects from his product."
> "…unless a relatively acute poison is involved an effect is not likely to be routinely traced to a
> causative agent in food or packaging."

**(3) A partial concession — then the demolition of the "single cut-off":**
> "Can we assume that any component of a food container that does not exceed a certain amount, say 0.2%,
> will always give an insignificant amount (less than 0.1 p.p.m.) in the total diet and therefore be safe?
> If we eliminate the pesticides, heavy metals, and carcinogens…, this generalization has considerable
> appeal and some reliability on the basis of 'published' toxicity data."
> "This generalization does not take into consideration whole meals in packaged materials, the use of an
> ingredient in [more] than one class of packaging material, and the additive effect of similar
> ingredients. Nor does it account for the food faddists… [or] that thick containers, such as plastic
> bottles, contribute far more to the food than plastic coating of a few mils thickness…"

**(4) The "published data are biased" methodological point + specific counter-examples:**
> "Long-term feeding studies as reported in the literature may not be representative of the total
> scientific knowledge on all possible and useful food packaging chemicals."
> — Some plasticizers extract into fatty foods at **100 p.p.m. or more**; FDA has had requests for
>   **triorthocresyl phosphate**; sensitizers like **toluene diisocyanate** and epoxy curing agents
>   are objected to **"even at levels below 0.2%"**; monomers like **acrylamide** are tightly restricted.

**(5) Conclusion:**
> "…more and more segments of our population are exposed to greater amounts of packaging chemicals. Some,
> including babies, may have their entire diet exposed… Therefore, I believe… that industry and FDA must
> know the relative safety of each migrant."

## The floor exchange that followed (verbatim report)
- **On migration evidence:** asked if there's evidence of migration greater than Frawley presented,
  Summerson "does not know"; Frawley said there is, but most substances do not.
- **On contrary studies:** asked whether any two-year studies "did not gibe with Frawley's," Summerson
  said there are **"plenty"** — "but not necessarily on known packaging components."
- **The survivorship-bias charge:** *"Summerson said that Frawley's sources 'bias the data,' since
  highly-toxic substances are not found in two-year studies, because the animals do not survive for two
  years."* (A sharp methodological hit: Frawley's 220-study database could only contain compounds
  non-toxic enough to *complete* a 2-year study.)
- **A prescient counter-example:** Frawley claimed nothing but pesticides/heavy metals is toxic at ≤40
  ppm; **FDA's Dr. Joseph McLaughlin, Jr. disagreed, "pointing to acrylonitrile."** (Acrylonitrile —
  the monomer in SAN/ABS food-packaging plastics — was later classified a probable human carcinogen.)
- **The split-the-difference signals:** Summerson — *"Somewhere in between will be a position that both
  Dr. Frawley and FDA can live with"*; Delmore (FDA) — the two sides are "not too far apart"; Heckman
  (SPI) — a "middle ground" is "a good place to start." Frawley — it's "just a matter of degree."

---
*To obtain Summerson's FULL paper: the proceedings/transcript of the FDA National Conference on Indirect
Food Additives (Feb 13–14, 1968) — not located online; try the FDA History Office, Food Drug Cosmetic Law
Journal 1968, or Food Chemical News' complete conference coverage (this clipping is pp. 5–7 only).*