John Forester, Petitioner, v. Consumer Product Safety Commission of the United States, Respondent.southern Bicycle League and James J. Berryhill, Petitioners, v. Consumer Product Safety Commission of the United States, Respondent, 559 F.2d 774 (D.C. Cir. 1977)

Petitioners seek review of regulations issued by the Consumer Product Safety Commission (CPSC) pursuant to the Federal Hazardous Substances Act (FHSA), 15 U.S.C. §§ 1261-74 (1970 & Supp. V 1975). These regulations establish standards for the manufacture and labeling of most bicycles, and classify bicycles not complying with the standards as banned hazardous substances.¹ 

The challenged regulations have had a lengthy and convoluted gestation period. Study of the risk of injury presented by bicycles was commenced in 1971 by the Bureau of Product Safety (BPS) of the Food and Drug Administration (FDA), Department of Health, Education and Welfare (HEW). CPSC Brief at 6. BPS considered submissions from various bodies, including the American National Standards Institute and the International Standards Organization (ISO), CPSC App. 378-401, and the Bicycle Manufacturers Association of America (BMA), R. 21, CPSC App. 363-377. Contracts for tests concerning bicycle design and safety were let to the University of Iowa, CPSC App. 473-527, 538-64, and Cornell Aeronautical Laboratories, CPSC App. 528-37, 565-761, and BPS analyzed bicycle-related injury data available from its National Electronic Injury Surveillance System (NEISS), which collects data from selected hospitals. CPSC App. 265-362.

On May 10, 1973, FDA published proposed regulations classifying "all bicycles intended for use by children of less than 16 years of age, whether for recreational, utility, or any other purpose" as banned hazardous substances under § 2(f) (1) (D) of the Federal Hazardous Substances Act, 15 U.S.C. § 1261(f) (1) (D) (1970), except those complying with the requirements of the regulations. Comments on the proposed regulations were solicited. 38 Fed.Reg. 12300 (1973). On May 14, 1973, responsibility for the administration of the FHSA was transferred by the Consumer Product Safety Act (CPSA), Pub. L. 92-573, 86 Stat. 1207, § 30(a) (Oct. 27, 1972), 15 U.S.C. § 2079(a) (Supp. V 1975), from the Secretary of Health, Education and Welfare to the Consumer Product Safety Commission. The Commission assumed the responsibility for the proposed bicycle regulations, and analyzed some 267 comments upon the proposed regulations. CPSC App. 84-100.

The CPSC published "final" bicycle regulations on July 16, 1974. 39 Fed.Reg. 26100 (1974). This set of regulations applied to all bicycles, not only those "intended for use by children." An extensive discussion of comments received and Commission responses appears in 39 Fed.Reg. at 26100-05. The regulations were to become effective on January 1, 1975. After the receipt of 50 more comments on the regulations, including a number from manufacturers, the Commission announced that a public meeting would be held concerning the regulations on September 9, 1974. 39 Fed.Reg. 31943 (1974). Because many of the comments indicated that an effective date of January 1, 1975, would not allow sufficient time for product design, testing and retooling, and because it desired additional time to consider additional amendments, the Commission elected on December 16, 1974, to suspend the effective date for an indefinite period. 39 Fed.Reg. 4356 (1974).

Further amendments and comments were published in January, 1975, and comments again solicited from all interested parties. 40 Fed.Reg. 1480 (1975). Additional public meetings were conducted on the subject, 40 Fed.Reg. 2581 (1975). On June 16, 1975, the Commission published its reactions to requests for changes, together with extensive proposed changes to the regulations. 40 Fed.Reg. 25181 (1975).

Another set of final regulations was published on November 13, 1975, again detailing its reactions to public comments on the proposed regulations. 40 Fed.Reg. 52815, 52828 (1975). The effective date was established as May 11, 1976, except for several sections that were to become effective November 13, 1976. Following further minor amendments,²  the regulations are now in effect in their entirety.

Shortly after the initial publication of these regulations, several manufacturers, groups, and individuals petitioned for review. The petitions were consolidated before this court on April 7, 1975. Since that time, as amendments have been made to the proposed regulations, all of the petitioners who are manufacturers or trade associations have voluntarily dismissed their petitions.³ 

The remaining petitioners are John Forester, an individual who has written and taught on the subject of bicycle design and safety (pro se), the Southern Bicycle League (SBL), an organization of adult bicycling enthusiasts, and James J. Berryhill, an interested individual.⁴  They attack the regulations on the grounds (1) that the CPSC lacks authority under the FHSA to issue regulations of the sort promulgated; (2) that the regulations were issued in violation of proper administrative procedures; and (3) that substantive provisions of the regulation are unsupportable.

Petitioners first argue that, for several reasons, the Commission was without jurisdiction under the Federal Hazardous Substances Act to promulgate the challenged regulations.

The FHSA prohibits "the introduction or delivery for introduction into interstate commerce" of "hazardous substance(s)," 15 U.S.C. § 1263 (1970), and provides for criminal penalties, 15 U.S.C. § 1264 (1970), and seizures of banned products, 15 U.S.C. § 1265 (1970). The Child Protection Act of 1966, Pub. L. No. 89-756, 80 Stat. 1303, § 3(a), 15 U.S.C. § 1261(q) (1) (A) amended the FHSA definition of "banned hazardous substance" to include

any toy or other article intended for use by children, which is a hazardous substance, or which bears or contains a hazardous substance in such manner as to be susceptible of access by a child to whom such toy or other article is entrusted . . .

The Child Protection and Toy Safety Act of 1969, Pub. L. No. 91-113, 83 Stat. 187, § 2(a), 15 U.S.C. § 1261(f) (1) (D) (1970), added to the definition of "hazardous substance"(a)ny toy or other article intended for use by children which the Secretary by regulation determines, in accordance with section 1262(e) of this title, presents an electrical, mechanical, or thermal hazard.

The 1969 amendments also clarified the meaning of "mechanical hazard."⁵ 

We do not understand the petitioners to question the authority of the Commission to promulgate regulations under these sections of the FHSA in proper circumstances.⁶  Neither do the petitioners challenge the Commission's jurisdiction under the FHSA over bicycles "intended for use by children."⁷  They do, however, argue that the Commission is without power under the FHSA to promulgate "design" regulations, that "adult" bicycles cannot be regulated under the FHSA, and that in any event bicycles intended for use by adolescents and adults are subject under the FHSA only to labeling requirements.

The SBL and Berryhill contend that the FHSA empowers the Commission only to ban hazardous substances outright, and does not permit the promulgation of "exhaustive regulations" dictating the characteristics that bicycles must possess. SBL Brief at 1-5. Petitioner Forester recognizes that the power to ban hazardous substances necessarily includes the power to define the characteristics being banned, but argues that the Commission is empowered to promulgate only "negative" or performance regulations that define hazards to be avoided through performance tests that leave design decisions to the discretion of manufacturers. The regulations issued by the Commission, he argues, contain impermissible "positive" or design requirements.⁸  Neither petitioner cites statutory language, case law or legislative history on this issue, but both point generally to the "intent" of the Act.

If the SBL is arguing that the Commission must ban all bicycles if it is to act at all, this petitioner is plainly in error. The power to define hazardous substances by regulation reasonably includes the power to make distinctions between products within a single product class. The 1969 Senate Report expressly recognizes this, and singles out bicycles as a specific example:

There are numerous toys and other articles on the market which can cause personal injury because of their mechanical aspects but would probably not as a class be banned. Bicycles are one example.

S.Rep.No. 91-237, 91st Cong., 1st Sess. 7 (1969).

There is, moreover, no support for Forester's argument that the Commission may issue only "negative" regulations. A limitation to performance standards cannot be inferred, as petitioners apparently suggest, from the fact that the FHSA provides generally for the proscription of "hazardous" products. Both design and performance standards can clarify the statutory definition of " mechanical hazard." The Senate Report accompanying the 1969 FHSA amendments implicitly recognizes that design requirements may be imposed in appropriate circumstances.⁹  In the Consumer Product Safety Act of 1972, Pub. L. No. 92-573, 86 Stat. 1207, 15 U.S.C. §§ 2051-81 (Supp. V 1975), which also amended the FHSA, Congress expressly permitted the use of design standards, although it expressed a preference for performance standards.¹⁰  Had Congress intended to prohibit the use of design standards under the FHSA, it could have included language to this effect in the 1966, 1969, or 1972 amendments. It chose, however, not to do so.

A regulation is presumed to be valid unless it is shown by those attacking it to be contrary to law. Hoffenberg v. Kaminstein, 130 U.S.App.D.C. 35, 36, 396 F.2d 684, 685, cert. denied, 393 U.S. 913, 89 S. Ct. 235, 21 L. Ed. 2d 199 (1968). The interpretation of a statute by the agency charged with its administration is entitled to "great deference" by the courts. Udall v. Tallman, 380 U.S. 1, 16, 85 S. Ct. 792, 13 L. Ed. 2d 616 (1965). The petitioners have not provided any justification for reading the FHSA more narrowly than the CPSA on this issue. We agree with the Commission that where in its sound discretion it determines that a design standard will most effectively protect the public from product hazards, it has authority under the FHSA to issue such a standard.

Petitioner Forester also argues that the Commission's jurisdiction under the FHSA is limited to regulation of "any toy, or other article intended for use by children," 15 U.S.C. § 1261(q) (1) (A) (1970), and that it is therefore precluded from regulating adult bicycles under that Act.¹¹ This language, he contends, empowers the Commission to regulate only articles intended for use by children, not articles that children might use. He concludes that "(i)t is easy to determine which bicycles are intended for use by children: these are small and primarily cheap." Forester Brief at 14.

The Commission found the distinction between adult and children's bicycles more difficult to draw, and argues that the scope of its regulations is justified by the fact that all bicycles except a narrow excluded class are "intended for use by children." The FDA's initial proposed regulations, 38 Fed.Reg. 12300 (1973), would have applied only to "bicycles intended for children of less than 16 years of age . . . ." The Commission found, however, that almost half of the 267 comments filed in response to the proposal questioned the means to be used to distinguish regulated from other bicycles. 39 Fed.Reg. 26100 (1974); CPSC App. 84-100. The Commission considered evidence indicating that many teenaged children, and some younger ones, are as tall as an adult, CPSC App. 101-25; and that in one sample, 22 percent of all persons injured while riding bicycles with wheel sizes of 25 inches or greater were children aged between 7 and 15. CPSC App. 159-258. The Commission also noted that advertising showed full-sized bicycles virtually indistinguishable in design and price marketed for both adults and children. CPSC App. 129-35. In its July 1974 publication of the first set of "final" regulations, the CPSC omitted the limitation to bicycles "intended for use by children under 16 years of age" and extended the scope of the regulations to all except "track" and "one-of-a-kind" bicycles as defined. 39 Fed.Reg. 26100 (1974).¹²  The Commission explained the reason for this change:

The Commission is aware that a large percentage of bicycles produced, particularly in recent years, are light-weight, relatively expensive and sophisticated bicycles which are bought by adults for commuting, touring, and other recreational purposes. However, these same bicycles can be, and are used by children and adolescents. It is clear there is no precise way of distinguishing between those bicycles intended exclusively for adults and those intended for children as well as adults. Neither the manufacturer nor the retailer can accurately predict who the subsequent user will be, nor can the seller predict whether the adult purchaser will be the exclusive user or whether the purchaser will give the bike to a child or share it with a child. Indeed, the bicycle may be purchased exclusively for adult use, and when a child in the family becomes physically able to ride it the use may change. Moreover, an adult purchaser may subsequently sell the bicycle to a parent for a child's use.

39 Fed.Reg. 26100 (1974). Before this court, the CPSC states its position:

This change involved wording rather than substance because the Commission found that bicycles intended for children under 16 years old include all bicycles except those specifically excepted from the regulations.

CPSC Brief at 13.¹³ 

The question to be resolved is whether a product like the bicycles in question that is used both by adults and by children is subject to regulation under the FHSA. Only one reported decision has considered a similar question under the FHSA, and the FDA in that case was considerably less aggressive in its regulatory reach. In R. B. Jarts, Inc. v. Richardson, 438 F.2d 846 (2d Cir. 1971) (Friendly, J.), the petitioner, a manufacturer of lawn darts called "Jarts," challenged a regulation banning all lawn darts except those which bore labeling indicating that they were not for use by or near children and which were not sold in toy stores or departments. The court noted that although the 1969 Senate and House Reports, S.Rep.No. 91-237, 91st Cong., 1st Sess. 5 (1969); H.R.Rep.No. 91-389, 91st Cong., 1st Sess. 9 (1969), both contain an extensive list of children's toys to be covered by the amendments, the Senate Report expressed concern that the term "toy or other article intended for use by children" not be read "so broad(ly) as to give the Secretary of Health, Education and Welfare the authority to prescribe design criteria for all products to which children have access." S.Rep.No. 91-237, 91st Cong., 1st Sess. 5 (1969), quoted in 438 F.2d at 852. The Second Circuit summarized this consideration:

It is plain enough that such articles as rattles, dolls, stuffed animals, nursery and playground equipment, and infants' carriages and strollers, which are intended only for use by children, fall within the statutory term. On the other hand, despite the mention of sports equipment in the Committee reports, the mere fact that a child might get hold of an article of that character which is kept in the home but is intended solely for use by adults, such as a regulation size golf-club or rifle, would scarcely make the statute applicable. Between these ends of the spectrum is a case like this where the article is intended primarily for use by adults but also for use by children when playing with them.

438 F.2d at 852. The court declined to hold that the Secretary could ban all Jarts outright, including those labeled and sold for use by adults. Instead, it upheld the Secretary's power only to the extent of the regulations actually issued, and reserved the broader question for a later case. 438 F.2d at 854. As the Commission notes, however, the court did uphold regulations governing the labeling and marketing of Jarts intended for adult use. CPSC Brief at 15 n. 4.

Since the Commission in the present case has promulgated regulations banning all bicycles not in conformance with them, and has not relied upon labeling or other techniques to distinguish "adult" from "children's" bicycles, we are faced with a variant of the question reserved by the Second Circuit in Jarts.

We share the Jarts court's concern that the Commission's power under the FHSA to regulate items for adult use should not be construed too expansively. The Commission reads Mr. Forester's argument too broadly when it states that he would have the court find the CPSC to lack jurisdiction under the FHSA over any item which "adults might use," CPSC Brief at 17. The language and intent of the FHSA turn upon intended use. As Forester seems implicitly to recognize, the relevant intent is like that in tort law a result is intended if it is a reasonably foreseeable result of one's actions.¹⁴  Incidental use of an item by adults would not deprive the Commission of jurisdiction to regulate the item under the FHSA, and incidental use by children would not create such jurisdiction.

Determination of such "intent," however, is vested in the sound discretion of the Commission. This is particularly so "(b)etween the () ends of the spectrum," where both "Jarts" and full-sized bicycles fall. In view of the evidence considered by the Commission in this case, we cannot conclude that it abused its discretion or acted contrary to law in determining that all bicycles except those excluded from the regulations are "intended for use by children."

Petitioner Forester also argues that the Act limits the Commission to establishing labeling requirements for bicycles intended for use by adults and children of sufficient maturity to understand and heed the instructions. Forester Brief at 14-16. He apparently relies upon the proviso to 15 U.S.C. § 1261(q) (1) (1970), which states:

That the Secretary, by regulation, (i) shall exempt from clause (A) of this subsection articles, such as chemistry sets, which by reason of their functional purpose require the inclusion of the hazardous substance involved or necessarily present an electrical, mechanical, or thermal hazard, and which bear labeling giving adequate directions and warnings for safe use and are intended for use by children who have attained sufficient maturity, and may reasonably be expected, to read and heed such directions and warnings . . . .

The Jarts court found it unnecessary to resolve a similar argument based upon this proviso. Instead, the Second Circuit found that if it did apply to Jarts on the ground that they "necessarily present (a) . . . mechanical . . . hazard," the regulations under review in that case did not transgress the proviso. 438 F.2d at 854. In the present case, the Commission has in part required labeling and instructions,¹⁵  but it is apparent that if the proviso is applicable the design and performance requirements are improper insofar as they apply to bicycles intended for adult and adolescent use.¹⁶  The Commission argues, however, that the proviso was meant to apply only to items such as chemistry sets that must contain hazardous substances if they are to perform their purposes at all. Bicycles, in contrast, can perform as bicycles even after hazards (at least those hazards identified by the Commission in the regulations) have been eliminated by redesign. CPSC Brief at 27-28.

We conclude that the Commission is free in its sound discretion to adopt design and performance requirements dealing with hazards that are susceptible to elimination by these means. The precise outline of the proviso must be determined only (1) where the design or performance regulations would seriously compromise the function of the article regulated, or (2) where the Commission seeks to ban a product outright rather than permitting its sale with warnings. The Commission has not sought to ban bicycles outright. A determination that the regulations issued do not seriously compromise the function of bicycles underlies the decision to promulgate the regulations. We consider Forester's challenges to individual provisions of the regulations below. For the present, we find that it is not contrary to law for the Commission to rely in part upon design and performance requirements affecting bicycles for persons of sufficient maturity to understand and follow instructions.

The SBL and Berryhill argue that the promulgation of final regulations affecting all bicycles after publication of proposed regulations covering only bicycles "intended for children" denied them due process of law. It is unclear why these petitioners chose to rely upon the Constitution rather than upon the specific provisions of the Administrative Procedure Act (APA) dealing with rulemaking.¹⁷  The APA, 5 U.S.C. § 553(b) (1970 & Supp. V 1975), provides that "(g)eneral notice of proposed rule making shall be published in the Federal Register," and that this notice shall include, inter alia, "(3) either the terms or the substance of the proposed rule or a description of the subjects and issues involved." We will consider the argument as one based upon the APA, since the statute provides at least as much protection as does the Fifth Amendment.

Section 553(b) does not require that interested parties be provided precise notice of each aspect of the regulations eventually adopted. Rather, notice is sufficient if it affords interested parties a reasonable opportunity to participate in the rulemaking process. See, e. g., South Terminal Corp. v. Environmental Protection Agency, 504 F.2d 646, 656-59 (1st Cir. 1974); California Citizens Band Assn. v. United States, 375 F.2d 43, 48-49 (9th Cir.), cert. denied, 389 U.S. 844, 88 S. Ct. 96, 19 L. Ed. 2d 112 (1967); Logansport Broadcasting Corp. v. United States, 93 U.S.App.D.C. 342, 346, 210 F.2d 24, 28 (1954); Willapoint Oysters, Inc. v. Ewing, 174 F.2d 676 (9th Cir.), cert. denied, 338 U.S. 860, 70 S. Ct. 101, 94 L. Ed. 527 (1949). The Commission argues that it is unrealistic for petitioners to contend that they were not provided sufficient notice of the "subjects and issues involved" in the rulemaking proceeding, including the potential scope of the regulations. Two hundred sixty-seven comments were received in response to the May 1973 publication of proposed regulations, and of these over 100 were directed to the means to be used to distinguish regulated from nonregulated bicycles. Petitioners themselves were among those who responded.¹⁸  The change in the definition of bicycles covered by the regulations was a clarification rather than a change of substance, continues the Commission, since it concluded that all except the excluded bicycles were in fact "intended for use by children of less than 16 years of age." CPSC Brief at 13, 47. Thus, the Commission urges us to hold that the variance between the 1973 proposal and the final regulations is not beyond that permitted by § 553(b).

We need not, however, reach this question. Instead, we find that sufficient notice that the Commission intended to regulate substantially all bicycles was published in the Federal Register subsequent to May 1973 and prior to the publication of the November 1975 regulations now in effect. The July 1974 regulations, suspended in December 1974, provided explicit notice of the eventual scope of the November 1975 regulations, and in fact prompted at least 50 additional comments and several public meetings. A number of notices were published during the interim period concerning the regulations, including the December 1974 suspension order and major sets of proposed amendments published in January and June 1975. In each published notice a caption indicated that the proposed regulations applied to "Bicycles" and the earlier publications defining the scope of the proposed regulations were cited. In the January and June 1975 notices a textual introduction explained that the regulations governed "certain bicycles" or "any bicycle" and the history of the regulations was explained. Interested parties thus by incorporation had constructive notice of the scope of the regulations being proposed and an adequate opportunity to participate in the rulemaking.¹⁹  This cured any defect in the notice provided by the May 1973 proposed rules. We therefore do not decide whether the variance between the May 1973 proposal and the July 1974 final regulations exceeded the bounds of § 553(b).²⁰ 

Petitioner Forester also assails sixteen specific provisions of the regulations, arguing that the Commission has not shown that each will eliminate a hazard that has caused a significant proportion of cyclist injuries. At the threshold, he has misread both the requirements of the FHSA and the relevant standard of review.

The Commission is empowered by the FHSA to determine that an article presents a mechanical hazard if it "presents an unreasonable risk of personal injury or illness." 15 U.S.C. § 1261(s) (1970). There is no indication in the language of the Act or its legislative history that the Commission was bound to develop a precise "body count" of actual injuries that will be reduced by each regulatory provision. A statement by Arnold Elkind, Chairman of the National Commission on Product Safety, favorably quoted in the 1969 Senate Report, supports the contrary conclusion:

When your intelligence tells you that something will create an injury and that it seems conceptually clear that an injury will occur, it is primitive to wait until a number of people have lost their lives, or sacrificed their limbs before we attempt to prevent those accidents.

S.Rep.No. 91-237, 91st Cong., 1st Sess. 2-3 (1969). We conclude that no precise statistical showing is required.

The Commission is, however, permitted by the FHSA to regulate only mechanical hazards that present "an unreasonable risk" of consumer injury. This means that the Commission must determine (1) that the risk posed by the hazard is an unreasonable one, and (2) that there is a sufficient nexus between the regulation and the hazard it is designed to prevent. The requirement that the risk be "unreasonable" necessarily involves a balancing test like that familiar in tort law: The regulation may issue if the severity of the injury that may result from the product, factored by the likelihood of the injury, offsets the harm the regulation itself imposes upon manufacturers and consumers.²¹ 

We now consider the specific challenges made by petitioner forester.²² A. Protrusions, Excluded Area and Thread Length

Section 1512.4(e) of the regulations prohibits "protrusions" that exceed 5/16 inch in height, and are less than 1/8 inch thick or have an end radius of less than 1/4 inch with a major and dimension of less than 1/2 inch. Excluded from regulation are protrusions so located that they cannot be touched by an object having a radius approximating that of a lower leg (3 1/4 inches), § 1512.18 (b). Nonconforming protrusions may be covered with a protective cap complying with the standards, § 1512.4(e)-(f). Section 1512.4(h) imposes a similar limitation on exposed screw threads. Section 1512.4(g) prohibits all protrusions along the top tube of the triangular bicycle frame except control cables and certain cable clamps, but permits conforming protrusions within 3 1/2 inches of the handlebar stem. Forester attacks each of these provisions. He challenges the protrusion provisions on the ground that they have not been shown to be likely to significantly reduce cyclist injuries, and because they will prohibit the use of many useful devices such as rear axle adjusting screws, derailleur parts, pump pegs and other items. He also attacks the excluded area along the top tube on the ground that it permits the use of stem-mounted shift levers, which he considers to be extremely dangerous to the groin area of the rider.

In support of these standards, the Commission cites data indicating that 85% of all bicycle-related injuries are lacerations, contusions, abrasions, and fractures, CPSC App. 271, and argues that many of these would be reduced by limiting the parts of the bicycles that could inflict such injuries. Forester replies that the statistics are meaningless since virtually all bicycle injuries result when the bicyclist strikes some object, including the ground or another vehicle. Moreover, many of the prohibited parts are not in areas of the bicycle where a rider would be likely to contact them in normal use, and many permitted parts could cause injuries if struck with sufficient force.

We have been unable to find in the record any more specific statistics or studies than that cited here by the Commission. The Commission has evidently not considered the utility of specific items that will be prohibited by the regulations, although it chose to except from the prohibition two items that it found to be useful, cable clamps and stem-mounted shift levers. In the absence of more specific study of the effect of the prohibition, we cannot find rational a limitation defined in such broad terms. The regulation applies as it is presently drafted, equally to protrusions of high utility that are located where they would virtually never be contacted by either a rider or nonrider struck by a bicycle. Some of these items are available only on highly sophisticated bicycles. While the regulation permits the retention of such parts if protected by a cap that meets the protrusion and other requirements, some of the items prohibited may not be useful if so capped. While we intimate no opinion as to particular items that may or may not be prohibited,²³  we remand to the Commission for reconsideration and further documentation on the specific justification for the protrusion restrictions and their impacts upon other useful bicycle equipment.

The exclusion for protrusions located within 3 1/2 inches of the handlebar stem, however, has received sufficient consideration by the Commission that we cannot consider it irrational. The Commission has expressly rejected a request by manufacturers and importers that the excluded area be increased to five inches, finding that 3 1/2 inches is sufficient to allow the use of adequate stem-mounted shift levers. 39 Fed.Reg. 26101 (1974). The Commission thus determined that such shift levers are useful devices and that they could be permitted, as restricted, without an unreasonable risk of personal injury to cyclists or bystanders. As the Commission notes, some injuries result from bicyclists striking the handlebar stem itself. It could rationally determine that the hazard is not unreasonable within 3 1/2 inches of the stem. We conclude that this portion of the regulations must stand.

Forester assails § 1512.4(i), which requires that control cable ends be capped or treated to prevent unravelling, on the ground that the Commission has not shown that a "significant proportion of cyclist injuries" are caused by fraying control cable ends. The Commission replies that the purpose of the requirement is to protect against skin punctures by frayed cable ends during bicycle maintenance. Forester agrees that such minor injuries may result from frayed cable ends, but argues that these injuries are so insignificant that they cannot support regulation.

We find this provision to rest upon a rational basis. Although the injuries caused in this manner are minor, the means to be used for reducing the risk of injury is itself quite inexpensive and has not been shown to interfere in any way with the function of bicycles. On these facts, it was within the discretion of the Commission to determine that the risk of injury from untreated control cable ends is "unreasonable."

Forester does not dispute the Commission's evidence showing that defective brakes present a serious injury hazard, CPSC App. 136, 273-76, but he objects to several aspects of the Commission's bicycle brake regulations as irrational. He first challenges the stopping distance requirement and test, §§ 1512.5(b) (1); 1512.18(2) (v)-(vi), which establish a shorter minimum stopping distance for bicycles with a lower high gear ratio, and which allow longer stopping distances for test bicycles carrying heavier riders and loads.

Forester contends that the latter provision is irrational because increasing rider weight does not affect center of gravity and hence has no effect upon stopping distance, although he points to no data to support his conclusion. The Commission considered test data showing that total weight does in fact affect stopping distance, and that rider weight does affect center of gravity, CPSC App. 402-72, 762-93, 872-968. We conclude that the Commission's decision to allow a correction factor for weight, in view of this data, rests upon a rational basis.

Forester argues that the longer minimum stopping distance for bicycles with a high gear ratio that will produce a top speed under 15 miles per hour is irrational since virtually all bicycles are ridden in traffic and down hills, where maximum braking force is required. It is not unreasonable to assume, however, as the Commission apparently did, that bicycles capable of higher speeds will more frequently be ridden at such speeds, and that slower bicycles will be ridden at lower speeds in traffic and on hills, as well as under other conditions, than will faster models. Thus, the Commission could rationally conclude that different braking capabilities, based upon different top speeds, were necessary to avoid an unreasonable hazard.

Forester also suggests that the Commission erroneously dismissed "pitchover" as a hazard in its braking tests. The tests require that equal braking force, mechanically measured, be applied to both front and rear brake handles throughout the braking process. Tests in which pitchover occurs are invalidated. He contends that the specified procedure is an ineffective and dangerous means of braking a bicycle, as the most effective procedure involves heavier pressure on the front brake than on the rear, with pressure released at the point of front wheel lockup to prevent wheel lockup and pitchover. The Commission responds that an NBS report, CPSC App. 872-968, shows that front wheel lockup occurs more frequently where surface coefficient of friction is low, whereas pitchover occurs where the coefficient of friction is high. Thus, the two events do not always occur together, and few pitchovers occurred during the Commission's bicycle tests, 40 Fed.Reg. 52818 (1975). The specification of mechanically measured devices and uniform procedures specified for the brake test can be justified by the need for uniformity and fairness to competing manufacturers. Although these provisions may not represent the best available brake test, they cannot be termed irrational.

Forester next contends that § 1512.5(b) (4), which specifies that hand brake assemblies be "attached to the frame" by means of fasteners with locking devices, is irrational. Forester points out that the Commission, in response to a query, defined "brake assemblies" to include both brake levers and caliper assemblies, 40 Fed.Reg. 52818 (1975). He argues that hand brake levers are never attached to the "frame" at all, but rather to the handlebars. He argues as well that there is no need for locking devices since brake levers and calipers do not immediately fail if they become loose, but give adequate warning to permit tightening. He also argues, without citation of authority, that no bicycle now uses lock washers on brake handles, that better equipment does not employ such devices on caliper assemblies, and that lock washers might interfere with effective positioning of caliper assemblies. We do not find that the Commission's requirement for suitable locking devices, not limited to lock washers, to be irrational. The Commission could reasonably conclude that loose brake assemblies pose a hazard to bicycle riders, particularly those who do not carefully maintain their bicycles. It could likewise reasonably conclude that the cost of locking devices would be small, as would be their effect on brake performance. Moreover, the inartfulness of the Commission's use of the term "frame" does not render the provision irrational. A fair reading of the provision as a whole suggests that "brake assemblies" includes the brake levers and attaching brackets, and that therefore "frame" refers to all components to which brake parts attach, including the front fork and handlebars.

Petitioner Forester also assails the requirement of § 1512.5(b) (7) that brake pad material not melt or blister when subjected to a temperature of 250 degrees fahrenheit for 30 minutes. He argues, first, that this requirement subjects the pads to heat in excess of any that could be encountered in actual service conditions, and second, that the Commission subjects coaster brakes to no similar heat test in § 1512.5(c). He suggests that this discrimination may be motivated by the fact that coaster brakes, unlike caliper brakes, are largely or entirely domestically produced. The Commission responds that the pad test is designed to insure that brake pads will not glaze or deteriorate when subjected to high heat, and that the tests are designed to deal with the " unique failure modes" of the two systems. CPSC Brief at 37 n. 13.

There is no record support for Forester's suggestion that the differences in treatment of the two braking systems is motivated by economic chauvinism or protectionism.²⁴  On the other hand, the record appears devoid of any rational basis for the distinction between the two systems. The record reflects no consideration by the Commission of coaster-brake failure due to overheating, despite the submission of an article by Forester containing test results showing that this was in fact a serious problem. CPSC App. 1574-75. The Commission's consideration of heat problems was apparently limited to caliper brakes, CPSC App. 1532. We thus find no support for the Commission's statement that the difference in the regulations is based on the "unique failure modes" of the two systems. Without more, this distinction is irrational. This portion of the regulations is remanded to the Commission for further consideration.

Forester next challenges the requirement that handlebars be between 14 and 28 inches in width, § 1512.6(c). The Commission defends this provision on the ground that it is justified by anthropometric data, App. 116, that it coincides with the voluntary standards adopted by the ISO and BMA, CPSC App. 372, 1506, and that this standard is necessary to "provide optimum handling of a bicycle under stable conditions." CPSC Brief at 37 n. 13. Forester argues, based on his own expertise, that adult bicycles equipped with racing-style dropped handlebars typically have handlebar widths of from 12 to 16 inches, about the width of the shoulders of the rider. Widths for children, he asserts, should be narrower. This configuration, he argues, produces optimal control; moreover, excessively wide handlebars may restrict maneuverability and themselves present a safety hazard.

A study in the record indicates that loss of steering control is a significant cause of bicycle accidents, CPSC App. 136. Congruence of the regulation with the BMA and ISO standards suggests that the standard is a reasonable one that probably does not interfere with the use of most bicycles. We have, however, searched the record in vain for any data indicating that handlebar width was a factor in any of the accidents studied or that widths of less than 14 inches present a hazard of injury when used by persons of the proper size. The anthropometric data in the record for children concerned only standard handlebars, not racing or "dropped" handlebars. The Commission cannot justify regulations by reference to "optimum" control it is limited by the FHSA to regulation of unreasonable hazards. The BMA and ISO standards do not reveal supporting data. While such private standards may tend to show the reasonableness of similar Commission standards, they do not prove the need for such provisions. In the absence of any indication in the record that widths of less than 14 inches present an unreasonable hazard of personal injury, we must remand this portion of the regulations to the Commission.

Forester next challenges two of the requirements of § 1512.7(a), regulating pedal construction. There is little question, and Forester does not dispute, that foot slippage off pedals is a significant cause of bicycle accidents, see CPSC App. 136. He argues, however, that the requirement that pedals have right-hand/left-hand symmetry is not safety-related, since pedals are functionally always slightly asymmetrical because cranks are somewhat offset, and there has been no showing that lack of symmetry could contribute to foot slippage or other injury hazards. The Commission noted, in response to comments on the proposed regulations, that the provision related only to pedals themselves, not to the crank assembly. 40 Fed.Reg. 25183 (1975). While the Commission has collected no data bearing directly upon the question of symmetry, we do not regard it as irrational for the Commission to conclude that substantially asymmetric pedals could contribute to instability and foot slippage. Certainly this requirement does not disturb conventional bicycle design. We thus do not find this provision invalid.

Forester also challenges the requirement that "the tread shall be an integral part of the pedal construction to the extent that removal of the tread material would substantially destroy the pedal," § 1512.7(a). The rationale for this requirement is that it will deter the further use of pedals after the tread material is worn off, since such pedals will become difficult to use, and that pedals of integral tread design will not fail by the sudden loss of tread material, resulting in loss of control. 40 Fed.Reg. at 25183-84 (1975). Forester reads this provision to require the use of a pedal that is expensive to replace and that therefore may often be ridden without proper tread, that will not be harder to use without tread than a replaceable tread pedal, and that will be prone to catastrophic failure under stress, for example when the rider is sprinting. It is possible that a properly designed integral tread pedal would not fail catastrophically and would increase rider safety at little cost in terms of bicycle efficiency or repair expense. However, the court shares Mr. Forester's confusion concerning the nature of the pedal envisioned by the Commission. While it is not necessary for the Commission to promulgate a design standard, fuller explanation of the provision is necessary to establish its rationality. It is therefore remanded to the Commission for this purpose.

Forester next challenges the requirement of § 1512.8 that drive chains have a tensile strength of at least 1800 pounds. He argues that this standard is impermissible because accident data does not show that chain breakage is a current cause of a significant number of cyclist injuries, and therefore that chains of lesser strength do not present an unreasonable risk of personal injury.

This challenge is without merit. The Commission explained the reasons for adopting this standard:

The 1800 pound value was selected to coincide with the minimum value specified by the International Standards Organization (ISO) in their standard R606. Information provided to the CPSC indicates that chain breakage is not a current problem area because industry has adopted the ISO standard, which has specifications to allow for life-cycle degradation.

39 Fed.Reg. 26103 (1974). It is unnecessary for the Commission to present data showing that specific numbers of persons have been injured by a particular product defect, providing the standard otherwise rests upon a rational basis. See supra pp. 788-789. It is within the Commission's discretion to determine that failing drive chains could present a serious risk of personal injury to cyclists, and that adequate design strength must be provided to allow for wear over the life of the product. Forester has shown no way in which adoption of the 1800-pound standard will in any way interfere with effective bicycle design, or that the Commission has otherwise struck an improper balance between various characteristics that a bicycle should possess. While the Commission may not base its regulation solely upon the unsupported standard of a private trade group, the private ISO standard is evidence of the reasonableness of the regulation. We conclude that the standard adopted was within the sound discretion of the Commission.G. Protective Guards

Forester assails two protective guard requirements imposed by § 1512.9. The first, in § 1512.9(a), requires that bicycles with a single front and single rear sprocket have a chain guard covering the chain and 90 degrees of the drive sprocket after its intersection with the chain. This standard is designed to prevent the entanglement of clothing and body parts, for example, fingers or bare toes, in the chain and chainwheel. See 40 Fed.Reg. 52834 (1975); CPSC App. 1490-94.

Forester argues that this standard is unjustified because the risk of injury can be alleviated by instructions, e. g., to wear trouser leg clips and shoes while riding, and not to insert fingers into the chain and chain wheel. Nevertheless, the Commission did collect data indicating that such entanglement was a source of personal injury. The Commission is not required to proceed by instructions where product design can reduce a hazard without unreasonable effect upon product usefulness. The Commission could reasonably have concluded that younger children, who are least likely to follow instructions meant to reduce this sort of injury, would most frequently use single chainwheel/single sprocket bicycles affected by this standard. There is no similar guard requirement for more sophisticated derailleur bicycles. There has been no showing that chainguards impair the function of the bicycles to which this standard applies, or that the cost is unreasonable in relation to the harm the guards are meant to prevent. We conclude that this requirement is reasonable.

Next, Forester challenges the requirement of § 1512.9(b) that derailleur bicycles be equipped with a guard to prevent the drive chain from interfering with rotation of the rear wheel in the event of improper adjustment or damage to the derailleur. Such guards commonly consist of a small metal or plastic disc attached between the gear cluster and the spokes. Forester argues, first, that there is no significant hazard because the drive chain usually does not interfere with the spokes when the chain comes off the sprocket, and second, that derailleur guards are ineffective to reduce whatever hazard does exist.

The Commission has cited no engineering or accident data in support of its standard. Forester does admit, however, that drive chains do occasionally come off the rear sprocket and interfere with wheel rotation. The guard is not meant to prevent this, but to prevent damage to the spokes in the event the chain does come off. The Commission could reasonably conclude that the chain would be less likely to cause sudden interference with a smooth guard than with the unprotected spokes of the rear wheel, and that such a guard would not significantly interfere with bicycle design. We find nothing unreasonable about this requirement, although the Commission could have provided additional data in its support.

H. Tires.

Petitioner next challenges the requirement of § 1512.10 that the nominal inflation pressure be molded onto the sidewall of wired-on tires, and that the tire be able to withstand 110 percent of this pressure. The requirement, he argues, does not address an unreasonable risk of personal injury. Moreover, he contends, since cyclists may properly use different inflation pressures for different surface conditions or rider weights, a nominal pressure is meaningless.

The Commission's conclusion that improperly inflated tires pose an unreasonable hazard rests upon a rational basis. Over-inflation may result in improper seating of the tire to the rim, resulting in tire blow-off. It may also result in premature tire blowouts due to cracks and splits in the tire carcass. Under-inflation causes the tire to "bottom" against the rim when a bump is struck, which may pinch the tube, causing a blowout, or damage the wheel rim. Damage to the rim results in "grabbing" of caliper brakes, which may reduce braking effectiveness or upset the rider's equilibrium. See F. DeLong, DeLong's Guide to Bicycles and Bicycling 167 (1974), quoted in CPSC App. 1580. Forester disagrees with DeLong's conclusions concerning tire blow-off, but the Commission may rationally choose to accept the view of DeLong rather than that of Forester. The Commission could also rationally conclude that improper inflation poses an unreasonable hazard and that a recommended inflation pressure would serve to reduce this hazard. Knowledge of the optimal pressure will aid inexpert cyclists, and will not prevent those with special needs and knowledge from inflating their tires to different pressures. There is no suggestion that molding these figures onto the sidewalls of molded wire-on tires will increase the cost or reduce the efficiency of such tires. As a result of comments received, the Commission has expressly exempted from this standard tubular and nonmolded wire-on tires. These latter types are normally used only by expert cyclists and application of the standard to them is impracticable. See 39 Fed.Reg. 26103 (1974), CPSC App. 1561. We conclude that this standard is reasonable.

Forester objects to the requirements that there be no missing spokes, § 1512.11(a); that there be adequate clearance between the tire and fork or frame, § 1512.11(b); and that the rims retain the spokes when the spokes are side-loaded in accordance with the rim test, §§ 1512.11(c), 1512.18(j). Each of these standards, he argues, is not directed to an unreasonable hazard. These requirements are meant to reduce the likelihood of wheel collapse and of inner tube puncture by spokes, and to ensure proper caliper brake performance. CPSC App. 1534-35. Forester argues that since spokes work in tension, there is no evidence that a few missing spokes will contribute significantly to wheel collapse. He also argues that brakes work as well when the wheel is closer to the frame than permitted by the standard. Finally, he contends that bent wheels and missing spokes do not contribute to inner tube punctures, although he admits that spoke ends that have not been filed after tightening may puncture an inner tube.

Again, Forester does not appear to argue that the standards interfere with sound bicycle design. The standards apply only to new bicycles, and we do not understand Forester to assert that it is desirable to purchase a new bicycle with misaligned wheels and missing spokes. Rather, he argues that these defects do not present an unreasonable risk of personal injury. When the product characteristics being regulated have no countervailing benefits at all, and the cost of eliminating the defects is slight, the burden of supporting the regulation is not heavy. Forester admits that missing spokes may contribute, if only slightly, to wheel collapse in accident situations. He also admits that proper spoke tightness, together with filing of spoke ends, reduces the likelihood of inner tube puncture from spoke ends. It was not irrational for the Commission to act on similar conclusions, given the minimal effect of these provisions on desirable bicycle characteristics. Finally, it was not irrational for the Commission to conclude that some minimal clearance between tire and frame was necessary for proper braking, and its standard of only 1/16 inch could not interfere with proper bicycle design. We conclude that each of these standards is reasonable.

Sections 1512.13 and 1512.14 establish strength standards for the front fork and frame, to be applied through tests set out in §§ 1512.18(k) (1) and (k) (2).

Forester readily concedes that sudden failure of a front fork can pose a serious hazard to the cyclist. This is borne out by data in the administrative record, CPSC App. 1424-89. He argues, however, that the test fails to distinguish safe from unsafe forks since it tests only resistance to sudden force, not resistance to fatigue failure from long-term vibration or stress. The test, he contends, is not a safety standard but a durability test based on a BMA/6 standard intended only to ensure that children's bicycles would survive minor collisions or being ridden over a curb.

The Commission did begin development of a fork test with reference to a proposed BMA standard, but arranged for testing and analysis of forks by NBS, CPSC App. 630-708. The test adopted by the Commission incorporated extensive modifications of the original BMA standard, and was intended to ensure that fork failure occurred gradually, through bending, rather than suddenly through brittle fracture. CPSC App. 685. The Commission concluded that a static load test reasonably ensured the safety of forks under all conditions of use, and adopted a permissible degree of deflection that would allow compliance by light-weight 27 inch forks, CPSC App. 687, 1564. In view of the Commission's extensive consideration of this matter and its conscientious attempt to balance competing considerations of safety, expense and convenience, and interference with bicycle design, we cannot conclude that its ultimate determination was unreasonable.

Finally, Petitioner Forester disagrees with the reflectorization standards of §§ 1512.16 and 1512.18(m). In these sections, the Commission establishes requirements that each bicycle have ten reflectors: one facing forward; one facing to the rear; two on each wheel, facing to the sides; and two on each pedal, facing to the front and rear. The side reflectors may be affixed to the spokes or may be molded onto the tire sidewall. Reflectors are required to meet the tests of § 1512.18(n), and reflector mountings must withstand the tests of § 1512.18(m).

Petitioner and the Commission agree that nighttime visibility to motorists is extremely important to cyclist safety. Petitioner argues, however, that a mixed head-light-reflector system is a much safer system than the all-reflector system mandated by the CPSC. Petitioner's opening brief appears to advocate that the Commission should have adopted a requirement that bicycles be equipped with a front light and rear reflector. In his reply brief, however, he makes it clear that he advocates that the Commission adopt no standard at all on this matter, but leave nighttime protection to the motor vehicle codes of the states. A headlight and rear reflector on bicycles operated at night is presently required in almost all states. See annotation to Uniform Motor Vehicle Code Ann. § 11-107(a) (1967).

Petitioner argues specifically that reflectors are far less effective than a headlamp in the situation in which a bicycle and automobile are on paths at right angles to each other, since standard bicycle headlamps incorporate a lens providing some illumination to the side. In contrast, he argues, reflectors have an extremely narrow angle of visibility outside the centerline of the reflector. Thus, once the bicycle's reflectors in this situation become visible to the automobile driver, it is too late to take corrective action. Forester also argues that the Commission's adoption of an all-reflector system will mislead the public into believing that this system is sufficient to protect bicyclists, when in fact it is inadequate.

The Commission developed the all-reflector system through a balancing of cost, durability, and effectiveness of alternative systems. See 40 Fed.Reg. 25485 (1975). It concluded that an all-reflector system would be less expensive to install, would be less susceptible to malfunction and would require less maintenance than would a system involving lights. In addition, reflectors require no volitional act to render them effective, in contrast to lights. The Commission also determined that the reflector system would do little harm to the functioning of bicycles. Although it recognized that a light system would be somewhat more effective, it noted that the standard would in no way prevent bicyclists from installing lights. Id.

Forester's argument assumes that cyclists who ride at night would, but for the standard, purchase, maintain and use headlamps. The Commission could rationally have concluded that this was unlikely, and, that many unsophisticated or infrequent nighttime riders would otherwise do so without any protection at all. Although most states require use of headlights and rear reflectors when actually riding at night, there is in most jurisdictions no routine effort to enforce these requirements. Moreover, the regulation is not inconsistent with these state statutes. The Commission could also have rationally concluded that it should not, by federal regulation, require all bicycle purchasers, including those who seldom or never ride after dark, to pay the penalty in cost and weight of a headlamp system. Reflectors appear to provide a significant margin of added safety at a relatively small monetary cost and loss in bicycle efficiency. In view of the Commission's careful balancing of the relevant factors, we do not find this standard to be irrational.

Forester also challenges the reflector mounting and alignment standards of §§ 1512.16(d) and 1512.18(m). The former requires that the rear reflector be positioned at least three inches below the saddle in a location that will prevent it from touching the ground if the bicycle falls over. The latter requires that the reflector mount withstand a force of 20 pounds in three directions, including the direction in which the bicycle would be picked up by the reflector, without significant change in the reflector's orientation. Forester argues that the former requirement is irrational because the only available reflector location in a bicycle without fenders or a carrier rack is behind the seat where it would be obscured by a saddlebag. He contends that the latter standard is irrational because reflectors are not meant to be used as a handle by which to pick up bicycles. Moreover, he contends, even if a reflector is bent, a cyclist of sufficient maturity is capable of bending it back to the proper orientation when so instructed. Both of these contentions are without merit.

The Commission could rationally have concluded that, since reflectors perform effectively only when properly oriented, they should be located and mounted in such a way that they will resist disorientation caused by minor accidents or abuse. It was within the Commission's discretion to specify placement and mounting of reflectors that would serve this purpose on standard bicycles without optional equipment, such as saddlebags. First, it is by no means clear that all saddlebags will obscure a reflector mounted in accordance with the standard. Second, the regulations apply only to new bicycles sold to consumers; a cyclist who wished to add a saddlebag that in fact obscured the reflector could presumably move the reflector.

Judgment in accordance with this opinion.