When starting its first large project, the standard for bicycles, the Consumer Product Safety Commission (CPSC) started with a standard from the Bicycle Manufacturers Association (BMA), BMA/6. BMA/6 had been started when the BMA members produced largely children's bicycles and that standard was intended, by BMA, to apply only to bicycles intended for use by children. The standard says so itself: it "is not intended to include sidewalk, delivery, or competitive racing bicycles." At that time probably no member of the BMA built bicycles that were designed for adult use. Schwinn, which was the only large American maker of good bicycles for adults, was not a member of the BMA and its bicycles for adult use did not conform to BMA/6. In fact, were bicycles intended for adults built in conformity to BMA/6, there would have been practically no market for them, because the adult cyclists of the time bought bicycles that were designed for use by competent cyclists of any age and these customers would not tolerate the performance limitations that were built into BMA/6; many of the bicycles that they bought were either actual racing bicycles or bicycles built with racing materials and components, whether or not the users actually raced them.
Consequently, the CPSC standard became a standard for bicycles intended for use by children. In demonstration of that fact, the CPSC standard was indeed issued under the Federal Hazardous Substances Act, which allows only (for articles used outside the home) the issuance of standards for "toys or other articles intended for use by children," whereas the standard could have been issued under the Consumer Product Safety Act, which allows the issuance of standards that apply to articles used by either or both children and adults. The CPSC chose, for its own reasons, to issue the standard under the FHSA instead of under the CPSA, despite being urged by cyclists to use the CPSA.
The proposed standard caused an uproar in the cycling community, whose members feared that the standard would be applied to all bicycles instead of only to those for children. The CPSC received some hundreds of letters, many that were literally obscene, objecting to the application of the standard for children's bicycles to bicycles intended for, and used by, adults. Adults would not accept the performance limitations built into BMA/6. In consequence, a spokesman for the CPSC announced that the proposed standard applied only to bicycles intended for use by children, quoting the words from the Hazardous Substances Act: "toys or other articles intended for use by children." That announcement quieted the uproar.
Having gained time for the uproar to quiet down, the CPSC then officially found that practically every bicycle sold in America was intended for use by children. 
The logic was faulty. The CPSC argued only that because some bicycles of every model were small enough for full-grown children to use, whether or not the maker or the user intended that any child would use them, all bicycles of that model were intended for use by children. That's like saying that since some fourteen-year-olds are large enough to fit the cockpits of Indianapolis racing cars all Indianapolis racing cars are intended for use by children.
The result of these self-chosen actions by the CPSC is that the present standard has always been, and must be regarded as, a standard for bicycles that are intended for use by children.
At the time at which BMA/6 was being developed, the individual states required particular safety equipment for cycling at night. The traditional equipment was the front headlamp and the rear reflector, although some states had requirements for additional equipment as well. Despite these requirements, the Bicycle Manufacturers standard BMA/6 specified an all-reflector system to provide protection against the risks of car-bike collision when cycling at night. The precise origin of this system is unknown to me, but it appeared in the standard BMA/6. Although the BMA specified this system, it had never tested its system for effectiveness in its intended function. In other words, nobody knew whether this system would significantly reduce the risks of car-bike collision when cycling at night. While BMA/6 allowed the use of a headlamp or a taillamp, it did not require either one. However, it did state the following in the instructions that it required to be delivered with every bicycle: "For dusk or night riding, the bicycle should be equipped with a white light in front and a reflector in the rear."
The BMA advocated the all-reflector system because it was cheap. The BMA argued that the all-reflector system was appropriate because children should not be riding at night in any case, but that if they did this system could be operated by children, meaning that it required minimum maintenance and no operational skills. The BMA argued that the all-reflector system was effective because at least one of the ten reflectors was visible to a driver whenever his headlamp beams pointed at the bicycle.
The CPSC adopted the all-reflector system of BMA/6, perhaps with some optical improvements, as the required, permanently installed, system of nighttime protective equipment in the first proposed version of its standard for bicycles. This version caused cyclists to inform the CPSC, before 1974, that the all-reflector system would not provide the protection that headlamps provided. On 1 July, 1973, I submitted a notification to the CPSC in the form of a legal brief explaining the defects in its regulation. Pages 33 through 42 of that document describe the defects of the all-reflector system and describe the virtues of the headlamp and rear reflector system. The headings are: Reflectorization; Reflectors are technically inadequate for the task; There is no evidence for the effectiveness of the proposed system; The disadvantages of the scheme are overwhelming; Summary of reflector requirements; Alternate proposal for lighting equipment requirements. This is not something that one could easily miss. The typical example was the cyclist, equipped with the all-reflector system, riding on an arterial road protected by stop signs, say northbound. Any west-bound motorist approaching from a side street, and therefore facing a stop sign, must yield to the cyclist on the arterial street. The motorist's headlamp beams are shining westwards across the intersection, while the cyclist is still well south of the intersection. Because the motorist's headlamp beams don't shine on the cyclist's reflectors, the motorist cannot see the cyclist or his bicycle. The motorist starts from the stop sign and collides with the cyclist whom he has seen only for a fraction of a second before the collision. The CPSC was told that in many common car-bike collision situations reflectors could not notify motorists and that only headlamps could. The CPSC was also told that under no circumstances could reflectors notify pedestrians or other cyclists, who do not have headlamps, of the approach of cyclists. The CPSC was told that the only reflector that performed the required function was the rear reflector, because motorists from behind had the cyclist in their headlamp beams for plenty of time to avoid a collision.
To consider that this information was presented to the CPSC in an ambiguous form in which the contents were poorly stated and mixed in with other information is utterly absurd. The CPSC was properly and clearly informed of the defects and the dangers of its all-reflector system very soon after it issued the first proposed version of its regulation.
At this time, the CPSC may have had some doubts about the efficacy of its all-reflector system. After all, the information provided to the CPSC proved that in many common car-bike collision situations the motorist's headlamp beams did not illuminate the cyclist's reflectors until too late to avoid the collision. For whatever reason, at some time during this period some members of the CPSC observed a demonstration of the all-reflector system, presumably put on by the BMA and the reflector industry. The demonstration was held in the CPSC's parking lot. It consisted in having one or more cyclists riding in a small circle while in the beams of one or more sets of automobile headlamps. The observers from the CPSC, situated alongside the automobile headlamps, saw that at least one reflector was visible at every angle between the cyclist's direction of travel and the headlamp beams. That demonstrated the object of the designers of the all-reflector system: at every angle of illumination, there would be at least one reflector facing the source of the illumination.
While this demonstration was, to some extent, a test of the optical performance of the all-reflector system, it was not, in any way, a test of the efficacy of the all-reflector system in preventing car-bike collisions or bike-pedestrian collisions or bike-bike collisions. However, this demonstration appeared to satisfy the CPSC engineers as to the value of the all-reflector system, despite the fact that they had been informed in writing that there were many nighttime collision situations in which no headlamp beams would illuminate the bicycle's reflectors sufficiently early to avoid the collision. Perhaps the CPSC engineers believed that bicycles do not operate as vehicles and therefore don't need vehicular equipment. Perhaps the CPSC engineers believed that bicycles were toys that, at night, were played with in the roadway. Those assumptions ignore the question of how cyclists would actually operate and what equipment that style of operation would require, a question that the CPSC engineers never investigated. Perhaps the CPSC engineers didn't understand the criticism and analysis that had been presented to them. If the CPSC engineers had consulted any of the recognized authorities in traffic law or in traffic engineering, to discover why the state traffic laws required headlamps, they apparently didn't learn anything, because it is inconceivable that any such person would tell them that it is acceptable to operate at night without a headlamp.
Despite the information that the CPSC had received from me, the CPSC commissioners issued in the Federal Register their official finding that "The (all-reflector system) provides adequate visibility to motorists under lowlight conditions." Yes, the reflectors were adequately bright if the motorist's headlamp beams illuminated them, but no, that was not adequate to prevent car-bike collisions, let alone the other types of nighttime collisions that occurred.
At about this time a study of car-bike collisions in Toronto was published and the CPSC received a copy. Under the heading "Operation in Darkness" the study reports: "In 20 of the 28 cases involving nighttime riding, a bicycle with no forward or side lighting was in collision with a motor vehicle which approached from a frontal or side direction." The ratio of 20 out of 28 should have removed all doubt about the high frequency of the types of car-bike collision for which the appropriate safety equipment is a headlamp, those types that I had been describing to the CPSC. Since in the right-angle collisions in which the motorist has the duty to yield  the motorist's headlamps cannot shine on the bicycle's reflectors sufficiently early to avoid the collision, there should have been no doubt about the inadequacy of the all-reflector system.
This is the state of information when the CPSC issued its bicycle regulation that included the all-reflector system as the required system for nighttime traffic safety. There was no evidence whatever that the all-reflector system did what it was claimed to do: provide "adequate visibility to motorists under lowlight conditions," at least not under the conditions when car-bike collisions would be prevented. Even more telling, the evidence that did exist demonstrated that the all-reflector system could not prevent several very frequent types of car-bike collision. Furthermore, it was patently obvious that the all-reflector system could have no effect at all in bike-pedestrian collisions and in bike-bike collisions, nor could it tell cyclists how to stay on the road and where to ride on the roadway.
Furthermore, by this time Kenneth Cross had done his first study of American car-bike collisions.  While this study did not separate nighttime from daytime car-bike collisions, it demonstrated that 44% of car-bike collisions were of the right-angle type and that 26% were of the types that I had described as requiring a headlamp for prevention if they occurred at night.
The CPSC's decision to require the all-reflector system and to state publicly that it provided adequate safety for cycling at night, which is what the public thought that the CPSC's finding of adequacy meant, and presumably is what the CPSC commissioners thought that they meant when they issued it, not only had no justification whatever but was calculated to endanger the cycling population.
When the CPSC issued the final version of its regulation, I sued it.
So did Jeff Berryhill of the Southern Bicycle League. That is remarkable; those whom the regulation was intended to protect sued the issuing commission because the regulation endangered those it was supposed to protect. I sued on several grounds: the regulation was issued unlawfully, covering bicycles that were intended for use by adults under a law that allowed only the regulation of "toys or other articles intended for use by children;" several requirements were not related to safety at all; several others set design standards instead of specifying and prohibiting characteristics that were dangerous; several others were supposed to prevent invented accidents that had never been reported in the 100 years of cycling, and, the big one, the all-reflector system endangered cyclists. Jeff Berryhill stuck to the unlawful issuing argument. About a dozen manufacturers also sued the CPSC, but all they wanted was sufficient time to sell their outdated stock; they didn't care whether the bicycles that they sold were safe or dangerous, useful or toys. Once they got their extra time, they dropped out of the case.
Regulatory cases don't come to a trial of fact, only to a court of appeal. Issuing the regulation is presumed to be the result of a trial of fact in which the competing parties present their evidence and the regulatory commission makes an informed and unbiased decision about the requirements of the regulation.
Quite obviously the CPSC had not made an informed and unbiased decision about its bicycle regulation. In many matters beside that of nighttime protective equipment the CPSC's engineers had advanced arguments for requirements that they had either invented or had adopted from BMA/6, arguments that freshman engineering students ought to have known were incorrect. In the case of the nighttime protective equipment, all the evidence that the CPSC had was against the effectiveness of its all-reflector system. The only arguments in favor of it was that it was cheap and required no skill to operate.
The CPSC was not an unbiased trier of fact. The bicycle regulation was its first big task and if that regulation were to be sent back by the courts there would be serious political and professional consequences. As I was told, there was fear that staff would be fired if the regulation did not go through. Therefore, the CPSC had a very strong motive for defending its regulation, whatever might be the effect on cyclists. While the CPSC kept mentioning its duty to children, it was quite obvious that cyclists didn't count for much in the CPSC's calculations.
Here are the arguments that the CPSC advanced in defense of its all-reflector system, at least those arguments that can be expressed in logical form. I make this distinction because the CPSC's attorneys also produced arguments that have no logical basis and confuse, presumably deliberately, using no equipment at all with using the formerly standard equipment of headlamp and rear reflector. The issue is not between using no equipment at all and using the CPSC's all-reflector system, but between using a proper headlamp and a bright rear reflector and using the CPSC's all-reflector system.
I follow each argument with my comments about the validity of that argument.
1: "These requirements are designed to increase the range of detection of a bicycle at night and thus reduce the probability that a vehicle/bicycle collision will occur."
Using front and side reflectors instead of a headlamp reduces the range at which detection occurs to a distance at which collisions cannot be avoided. Using the CPSC-specified rear reflector, which is only 10% as bright as others which are available, reduces the detection distance from the rear.
2: "In deciding not to require lights the Commission considered several problems associated with lights. For example, required lights are more difficult to maintain in operating condition than reflectors. Batteries corrode, bulbs burn out, and generators fail. A bicycle with non-functioning lights can be ridden at night, but it may be an unseen target of motor vehicles. The consequences can be extremely serious if a child fails to maintain lights in operating conditions or forgets to turn them on. Permanently-mounted reflectors, on the other hand, require no activation and little or no maintenance on the part of the bicycle rider and will generally be effective for the life of a bicycle. The Commission has attempted to mandate requirements that will not only make newly-manufactured bicycles safe but will keep bicycles safe for children for as long as they are used."
This argument entirely ignores the question of whether the all-reflector system actually works. Sure, the reflectors retain their reflectivity for long periods, but the question of whether reflectivity is effective in preventing nighttime accidents is ignored. That's like arguing that tee shirts provide the appropriate protection against bullets because they are easy to wash.
3: "The Commission has evaluated the studies in the record on traffic patterns and injuries (Appendix 1103-1423) and concluded that illumination on both sides and on the front of the bicycle will reduce the occurrence of vehicle/bicycle collisions in other traffic situations, as well, such as where a bicycle crosses the path of a motor vehicle."
The only words about nighttime accidents in these 320 pages of the official record are the following. "There seems to be a need for more highly visible bicycles and riders. Application of retroreflective materials to a vehicle and/or rider and better lighting on bicycles and roadways may increase safety for night riders."
This quotation is from the Pedestrian and Bicycle Safety Study, Highway Safety Act of 1973, from the U.S. Dept. of Transportation, March 1975. This study was issued, even started, too late to have provided information about adopting the all-reflector system. The strongest argument that could be made by the CPSC is that while the CPSC didn't have the appropriate information to adopt the all-reflector system, information supporting its decision appeared later. That is, the statement about using this information to adopt the all-reflector system is a lie. However, the information does not support the all-reflector system. This statement fails to distinguish between reflectors and lighting, between on-bike lighting and roadway lighting, and it shows no knowledge of how nighttime accidents occur, knowledge that is absolutely necessary to make valid recommendations for preventing them. In short, the words merely suggest the need for research and provide no support for the all-reflector system.
The CPSC's statement is incorrect in another way also. The CPSC did possess a study about nighttime car-bike collisions, An Investigation of Rider, Bicycle and Environmental Variables in Urban Bicycle Collisions, from the Ontario Department of Transport, October 1970. This study appears in the official record at pages 1063 to 1100, outside the range of pages stated by the CPSC as covering the investigation of nighttime accidents. The entire passage about nighttime accidents is as follows. "Operation in Darkness. In 20 of the 28 cases involving night-time riding, a bicycle with no forward or side lighting was in collision with a motor vehicle which approached from a frontal or side direction at an intersection. The remaining cases involved overtaking man_uvres by the bicyclist or motorist. There were only 4 cases where rear lighting of the bicycle was operationally relevant."
Quite clearly, the only formal study possessed by the CPSC that included an analysis of nighttime accidents positively indicated the importance of car-bike collisions for which the bicycle headlamp is the appropriate countermeasure. Equally clearly, the CPSC concealed its possession of this study by directing attention elsewhere.
4: "The [CPSC's] Bureau of Engineering Sciences submits the following epidemiological and engineering rationale in support of the various provisions of the bicycle regulation. Requirements for Reflectors. BES has not recommended requirements for active lighting devices as these would tend to encourage night riding with a corresponding increase in exposure to the hazards. Additionally, such active lighting devices require maintenance, which is often neglected, resulting in even poorer visibility."
As a result of this prattle about protecting cyclists, those who choose or who must cycle at night are sent out with equipment that does not protect them from the hazards.
The CPSC included the all-reflector system in its preliminary issue of the regulation, based on the word of the bicycle industry as given in the standard BMA/6. This mistake is understandable, given the lack of formally stated knowledge about nighttime equipment at the time. However, the CPSC was then informed that this was a serious mistake. I provided the information, in clear engineering description, that only headlamps and not reflectors could provide protection in several important types of car-bike collision, as well as in bike-bike and bike-pedestrian collisions. The Toronto study confirmed that these types of car-bike collisions were a very high proportion of nighttime car-bike collisions. The first Ken Cross study of car-bike collisions demonstrated that these types of car-bike collisions were a very high proportion of all car-bike collisions, but without distinguishing daytime from nighttime.
At this point the CPSC should have reconsidered its advocacy of the all-reflector system and, at least, performed some simple tests. All that the CPSC had to do was to go to an unlighted intersection with a car and a bicycle equipped with headlamp, the CPSC's all-reflector system, and a bright rear reflector from an auto parts store. With the car facing west at the intersection, for example, the driver of the car could watch the approach of a cyclist from his left, first with the headlamp off and then with it on. The next test would be to determine the relative brightness of the bright rear reflector against the CPSC's deliberately dim one by driving behind the bicycle on a dark road. There is no question of what the test would have shown, because it has been done often enough by others.
Instead of either making a decision in accordance with the only engineering information about the performance of reflectors in traffic situations that had been given to it, or of investigating the validity of that information by performing very simple, cheap, and obvious tests, the CPSC chose to rely on rationalistic arguments about the value of its all-reflector system and about discrediting those who had provided the valid information.
The CPSC produced five arguments in support of its all-reflector system:
1: Reflectors increase the range of detection. False: headlamps are visible when front and side reflectors are invisible.
2: Lights are too difficult for regular use. However difficult lights may be for use, if they are the only way to prevent accidents then they must be used.
3: Bicycles are for children to use and maintain. False. This ignores adult use, particularly by adults who choose or who must cycle at night. This also ignores the fact that many aspects of bicycle maintenance must be performed by adults.
4: The CPSC based its choice on studies. False. The CPSC referred to no studies of nighttime accidents, the CPSC ignored the one study of nighttime accidents that it possessed, the CPSC ignored the studies that had been sent it in opposition to the all-reflector system, and the CPSC failed to make any studies of its own.
5: The CPSC chose not to require headlamps for nighttime cycling because to do so would encourage nighttime cycling. That is an immoral and unlawful argument because it endangers those who choose to or who must cycle at night.
The Court found for the CPSC. The Courts reasoning goes as follows: "The Commission developed the all-reflector system through a balancing of cost, durability, and effectiveness of alternative systems. 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. Although it recognized that a light system would be somewhat more effective, it notes that the standard would in now way prevent bicyclists from installing lights. 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; rather, that many unsophisticated or infrequent nighttime riders would otherwise do so without any protection at all. There is in most jurisdictions no routine effort to enforce (headlamp) requirements. 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. In view of the Commission's careful balancing of the relevant factors, we do not find this standard to be irrational."
That is the legal finding: the standard is valid because choosing to require the all-reflector system was not an irrational decision. But does it make engineering sense? That is not the criterion that courts use. However, consider the reasons by which requiring the all-reflector system was considered not irrational. The grounds were ease of use contrasted against not using anything at all. This was bolstered with the specious argument that the only other alternative to nothing at all would be requiring lights on all bicycles at all times, whether or not used at night. That is not a rational program for cyclist nighttime safety.
The Consumer Product Safety Commission Improvements Act of 1976 included the provision that no state or local government could have a regulation about some risk that was not identical with the federal government's regulation about that risk. The bicycle industry was very interested in getting that provision enacted. Once it was enacted, the Bicycle Manufacturers Association and Schwinn Bicycle Company went to the National Committee for Uniform Traffic Laws and Ordinances (NCUTLO) with the request that the Uniform Vehicle Code be changed to be identical with the CPSC's regulation regarding the all-reflector system. The attorneys for BMA and Schwinn's governmental relations director informed the NCUTLO that this was purely a formality, because the CPSC Improvements Act had already invalidated all state requirements that were not identical with the CPSC's regulation. If the NCUTLO did what the law now required, that act would make the changeover go more smoothly than if the BMA and Schwinn had to get each state in turn to recognize the change.
Since the federal government required the all-reflector system to prevent car-bike collisions at night, and since the states required headlamps to prevent car-bike collisions at night, in theory the CPSC's requirement of the all-reflector system invalidated all state laws requiring cyclists to use headlamps when cycling at night.
I argued before the committee that if they were to accept BMA's and Schwinn's request, they would be making motorists, their primary constituents, liable for nighttime car-bike collisions with bicycles that it was impossible for them to see. The argument carried and the NCUTLO overwhelmingly rejected BMA's and Schwinn's request.
However, the legal question remains: do the CPSC Improvements Act and the CPSC's bicycle regulation actually do what the law requires, invalidate state laws for the use of headlamps at night? Nobody knows the answer to that question, and if it gets litigated by parties with much at stake, it will probably end up in the Supreme Court because no lower court has the authority to make its opinion prevail in this matter.
The practically unanimous rejection of the all-reflector system by the nation's foremost committee of experts in traffic operations and traffic law should have alarmed the CPSC and given it cause for reconsidering its all-reflector system. However, no change was observable from outside.
Publication of Cross's second study of car-bike collisions  provided the first detailed study of car-bike collisions with the data classified by daylight and nighttime. By comparison with daytime data, some types of car-bike collision appeared to be much more frequent during night. These types are motorist exiting from side street, motorist turning left in front of the cyclist, wrong-way cyclist hit headon, and motorist overtaking the cyclist. The first two of these are those types that I and the Toronto study had informed the CPSC were very important and required a headlamp for prevention, and the third presents a physical situation that is very similar to the second. In 1983 I published a quantitative analysis of these data.
|Type of Car-Bike Collision||Percent|
|Motorist exiting from side street||47.2|
|Motorist turning left||22.3|
|Wrong-way cyclist head-on||9.5|
This analysis of Cross's data shows unequivocally the importance of the types of car-bike collision for which the headlamp is the primary preventive measure. In the same book I also made an estimate of the casualties incurred in these types of car-bike collision.
|Type of Car-Bike Collision||Casualties/yr|
|Front reflector substituting for headlamp||400||4|
This table shows the number of casualties incurred as a result of the failure
to use particular items of nighttime protective equipment, or, conversely, the
number that would be saved if that equipment were used.
These statistics should have compelled the CPSC to re-evaluate its advocacy of the all-reflector system, but there was no change.
Little happened over the next decade to affect the CPSC's opinion about its all-reflector system. I know of two car-bike collision cases in which the lack of a headlamp was the cause of the collision, but these were litigated with entirely different intent, largely because the attorneys involved did not understand the kind of cases that they were handling. One was litigated as a lack of bikelane case, and was successfully defended on the grounds that the cause of the collision was the lack of a headlamp rather than the lack of a bike-lane stripe.
However, the case of Johnson vs Derby Cycles  got to the heart of the controversy. Johnson was a high-school student with an evening job. He bought anew bicycle built by Derby Cycle Co., expressly to ride to and from his job, which ended at midnight. The bicycle was sold to him with the required all-reflector system. The bicycle shop did not advise him to use a headlamp. The bicycle safety booklet that Derby published and supplied to conform to the CPSC regulation informed the cyclist to use a headlamp for off-road cycling at night. Johnson had cycled when younger in Trinidad, but not recently. With less than one week of experience riding to and from work, he left work one evening. His return took him over a ridge of hills with an 8% descent. While he was on that descent, probably traveling between 30 and 35 mph on a road zoned for 35 mph, a motorist ascending the road suddenly turned left in front of Johnson. Johnson collided with the car and was severely and permanently injured in body and in brain.
Johnson's attorney sued Derby Cycles. During the litigation, Derby's officers stated that they relied on the CPSC's regulation for safety requirements, that they had no idea that the all-reflector system was unsafe, and that they did not know that state laws required the use of headlamps when cycling at night.
A survey of high-school students in New Jersey showed that about 85% believed that bicycles with only the all-reflector system were safe to ride at night, and that very few of those of them who rode at night used any other nighttime equipment.
Johnson won a verdict of $7 million, later reduced to an actual payment of about $2 million. The case attracted a very large amount of attention. The facts of the case reflect exactly the predictions that the CPSC had received twenty years before. That is, the all-reflector system could not perform the functions of the headlamp, but the presence of the front reflector misled users (in this case, both the cyclist and the bicycle manufacturer) into believing that it did, producing catastrophic injuries to the cyclist and similar financial burden on the manufacturer.
Derby Cycle Company is not the only company in the bicycle industry that, either intentionally or unintentionally, misleads the public about the safety of the all-reflector system. Consider the manual that Specialized Bicycle Components supplies with its headlamps. Specialized has a good reputation among cyclists for the design and quality of its products, which, despite the name of the company, now include complete bicycles as well as components. The example that I make of its manual for headlamps shows the extent to which members of the industry are forced, in self protection, to participate in the CPSC's policy of deceiving the public about the safety of the all-reflector system.
The Specialized manual contains paragraphs headed Warnings and Cautions. Under warnings, the manual includes useful safety instructions, such as carrying a spare bulb and the screwdriver or coin needed to change bulbs. The warnings also state:
"Wear reflective clothing. Make sure your helmet, pedals, ankles (or shoes), packs, and panniers have reflectors or reflective patches."
Under cautions is the specific paragraph:
"Specialized lights and lighting systems are not intended to replace C.P.S.C. approved reflectors. Never remove the original reflectors from your bicycle."
These statements are lies in that they tell the exact opposite of the truth. They say that reflectors provide a safety function that the headlamp or lighting system does not provide. Obviously, from the nature of reflectors that even the least educated understand, the safety function that reflectors provide must be protection against car-bike collisions. The truth is the exact opposite. The all-reflector system cannot provide the protection against car-bike collisions that it falsely pretends to provide. Only a headlamp can provide that protection for the great majority of car-bike collisions that occur from in front or the side of the bicycle.
In order to protect themselves against liability, the members of the industry, even if they know better themselves (and Derby claimed that it did not know better), have to participate in the public deception that is part of the CPSC's regulation with its requirement for the all-reflector system.
In June of 1994 the CPSC published its Bicycle Use and Hazard Patterns in the United States. This study drew immediate criticism from the best informed cycling transportation engineers in the nation. Among other things, the data in the study claimed that the average speed of adult cyclists was 1.2 mph. Lest that seem a typographical error, I repeat: 1.2 mph. This is not the place for general criticism of the CPSC's study. Suffice it to say that the study's many flaws in addition to those about the all-reflector system demonstrate that the CPSC's lack of competence in cycling transportation engineering has continued from the days when it initially adopted the all-reflector system.
The flaws in the study are concentrated in the subjects of time and place. Where simple counting is concerned, the study is more accurate. The study states the following about nighttime cycling:
"The use of bicycle headlights and reflective clothing should be encouraged. Night riders should also make sure that their bicycles are equipped with reflectors, as required by the CPSC bicycle standard."
"About 12 percent of bicyclists ride at least occasionally after dark. However, less than one-third of these nighttime riders use headlights or taillights." (Pg 4)
"Tail lights and headlights which, according to Ferguson and Blampied (1991), can substantially reduce the nighttime accident risk, were the most widely reported bicycle safety accessories. Tail lights and headlights were respectively reported on 20.6 percent and 14.5 percent of bicycles." (Pg 27)
"Riding during non-daylight conditions (i.e., dawn, dusk, and night) was a significant factor in the risk of injury for children. Non-daylight incidents were more common on major thoroughfares than in other locations. NHTSA data on pedalcyclist deaths involving motor vehicles suggest that night riding also may be a contributing factor in fatal incidents (14) [This 14 refers to the Fatal Accident Reporting System (FARS) data.]
"While it seems intuitively apparent that riding during dawn, dusk, or night would be riskier than at other times, it is possible that some people perceive reflectors as adequate protection at times when they may not be sufficient. It was reported that most bicycles were equipped with reflectors. However, it was beyond the scope of this study to determine the adequacy of the mandatory standard's reflector requirements.
"While specific conclusions could not be drawn about the use of bicycle lights during non-daylight incidents, less than eight percent of all bicycles involved in injuries were reported to have been equipped with lights (regardless of daylight conditions). Of those with lights, a few were involved in incidents that occurred at dawn or dusk while the light was not being used.
"Night riding may be an area deserving future information and education efforts (e.g. the need for bicycle lights, reflective clothing, etc.)." (Pg 78)
These are all the words about nighttime cycling and nighttime protective equipment that are in the study. This study is the first formal study of cycling safety by the CPSC since its involvement in issuing its standard. This study is a major effort with contributions from many different employees of several departments of the CPSC. The printed copy has 153 pages. This study must be presumed to at least summarize the CPSC's knowledge of the safety aspects of nighttime cycling.
Consider first the literature search aspects of the study. Does the CPSC know the knowledge that has been published in the field? The only scientific study to which the CPSC refers is one by Ferguson and Blampied. The CPSC asserts that Ferguson and Blampied state that "tail lights and headlights can substantially reduce the nighttime accident risk." Ferguson and Blampied made no analysis concerning this subject. They reported on a largely unsuccessful effort to persuade university students in New Zealand to use lights at night, based on Ferguson's and Blampied's belief about the virtues of lights. Ferguson and Blampied specifically cite a study by Atkinson and Hurst of car-bike collisions in New Zealand  as the scientific support for their belief. The Atkinson and Hurst study was published in Accident Analysis and Prevention, an American journal that should be in the CPSC's library. The CPSC should have read Atkinson and Hurst to learn the data that they presented about the risks of nighttime cycling.
The Atkinson and Hurst study specifically refers to Cross's second study of car-bike collisions as describing the method that Atkinson and Hurst followed. Cross's second study, to which I have repeatedly referred, is the landmark study in the field of car-bike collisions. The data in that study are the source of tables 1 and 2 given earlier in this paper. The CPSC apparently has never heard of Cross's work, even though it was commissioned by a sister department of the CPSC, the National Highway Safety Administration.
I think that I can claim, without boasting, that I am the world's foremost writer about cycling operation and safety. In the book Bicycle Transportation I wrote the analysis that included the tables 1 and 2 of this paper, a book published by the Massachusetts Institute of Technology in 1983. (Second edition in 1994) That book specifically cites the Cross study as the source of the data. That book also includes descriptions and diagrams showing why the all-reflector system cannot inform motorists of the presence of the cyclist in the types of nighttime car-bike collisions that Cross's data show to be most frequent. The information in the books is only a presentational refinement of the descriptions that I submitted to the CPSC in 1973. Considering that I was the principal objector to the all-reflector system when the CPSC first proposed requiring it, and that I sued the CPSC over this issue, and that MIT has published my subsequent work, one would think that the CPSC would have paid attention to my work, at least when it decided to investigate the subject of nighttime protective equipment.
Furthermore, most studies of car-bike collisions that have provided far less detail than did Cross also conclude that the car-bike collision rate at night, considering the exposure, is considerably higher than the daytime rate. This is also a standard conclusion about highway traffic of all types.
The evidence shows that the CPSC has remained ignorant of the basic studies that cover the field of nighttime cycling safety.
Now consider the CPSC's consideration of the data that it does know. It reports that less than one-third of cyclists out at night use either a headlamp or a taillamp. It reports that most bicycles are equipped with reflectors. It reports that headlamps are fitted to only 14.5 percent of bicycles, but taillamps are fitted to 20.6 percent of bicycles. These data would not matter at all if the all-reflector system actually was adequate, as the CPSC once officially found. However, in the light of the evidence the data are damning. The great majority of people obviously rely on the all-reflector system for nighttime safety. Of those who choose to use additional equipment, the largest number probably add more reflectors. Of those who choose to use lights, 50 percent more choose to fit a taillamp than choose to fit a headlamp. A rear-facing reflector is the only reflector of the CPSC's required ten that is properly positioned to perform a necessary safety function, because only it is in the motorist's headlamp beams early enough to prevent the collision. The front and side reflectors cannot provide that necessary safety function because they are not in the motorist's headlamp beams early enough to prevent the collision, to say nothing of their complete inadequacy in preventing collisions with entities that don't project headlamp beams before them. In other words, at least one-third of those people who choose to add lights make the wrong decision and add the light to the one reflector that is correctly positioned to operate while neglecting the other reflectors that cannot operate and whose failure to operate causes about 80 percent of the car-bike collisions that are caused by darkness. This is surely evidence enough that people are dangerously misled by the policy of requiring the all-reflector system and suffer numerous injuries and deaths as a result.
Now consider the CPSC reasoning about nighttime protective equipment.
In the Overview section the CPSC recommends that headlamps and reflective clothing be used at night, with a special instruction that "Night riders should also make sure that their bicycles are equipped with reflectors, as required by the CPSC standard."These words give the clear impression that the CPSC policy is to rely mostly on reflectors, both on the bicycle and on the person, for nighttime safety. The plain fact is that if highly reflective reflectors carefully installed on the bicycle can't alert the motorist because the headlamp beams don't illuminate them, then less reflective materials disposed according to the uncontrolled movement of clothing can't do so either, for exactly the same reason.
The CPSC concludes that nighttime cycling "was a significant factor in the risk of injury to children." Ah, yes, the CPSC regulation covers only "toys or other articles intended for use by children." Do children ride in such a way as to have a greater risk at night than do adults? That is what that sentence means to me. However, the statistics of those other studies that have considered the matter show that the risk is at least equal for adults because a greater proportion of cycling by adults is done at night, producing a greater proportion of total accidents to adult cyclists, than is the case for children.
The CPSC states: "While it seems intuitively apparent that riding during dawn, dusk, or night would be riskier than at other times, it is possible that some people perceive reflectors as adequate protection at times when they may not be sufficient." What has the time got to do with the problem? It is absurd to expect that the motorist approaching the stop sign at the cross street on my right will see my reflectors if the time is between 8 and 10 pm, but not at other times. How is anyone to know at which times to rely on reflectors and when not to? This is another of the explanations depending on magic rather than reason or science that keep appearing in discussions of cycling safety. Any system that is so unreliable should not be used when there is a more reliable system available.
The CPSC states that "it was beyond the scope of this study to determine the adequacy of the mandatory standard's reflector requirements." Then what was the purpose of the study? What should it have been?
It is important to remember that the initial controversies about the standard for bicycles, except for the controversy about the all-reflector system, largely discussed two aspects. The initial controversies largely concerned the proposed standard's prohibition of good bicycles, because it specified toy bicycles built for children by the American bicycle industry, which at that time ignored the production of good bicycles for either adults or children. When the standard was modified to allow many characteristics of good bicycles those controversies largely died down. The second lot of controversies concerned the absurdity and uselessness of many requirements. The requirements for handlebars suitable only for gorillas, for pedals that fell off when worn out, or that prohibited valve stems, pump pegs, derailleur adjusting screws, etc., were some of those that remained and were thrown out by the court of appeals. Others that were not thrown out were considered not irrational, although they were not functional. For example, the requirement and test for front fork strength does not measure front fork durability, which is the characteristic in question. Sure, failure of the front fork is very dangerous, but the test does not test for that type of failure. The requirement for the ability of the rim to resist increased tension from the spokes is not related to any accidents that have been reported in 100 years of cycling, and assumes a mode of wheel operation that is exactly the opposite of what occurs. In other words, most of the revised standard was criticized as useless in reducing deaths and injuries to cyclists.
This is born out by the results of the study. Studies made before the
standard came into effect showed that 8 percent of injuries may have been caused
by product failure, while the CPSC study itself concludes that, after the
standard has been in effect for almost 20 years, 10 percent are so caused. Other
studies show that most of the injuries associated with product failure are
caused by worn-out equipment that has been poorly maintained, and hence is
beyond any remedial action by the CPSC's standard. 
In short, the study could never have determined whether or not the CPSC's standard was working, because the data associated with the items that are regulated are so few and are so confounded with those many from other causes that they cannot be distinguished by such a study. As it happens, all that the study has shown, as far as the CPSC is concerned, is that the CPSC's standard is useless, as was originally maintained. The data in the study, inaccurate as they are, have demonstrated two other conclusions. The first is that questions of cyclist safety belong in the areas of traffic operations and traffic facilities. The second is that the CPSC is even more incompetent in these areas than are the present governmental employees, and every effort should be made to keep the CPSC out of these areas.
So far as the CPSC's legitimate responsibilities are concerned, the one requirement of the standard that was always criticized as being acutely dangerous (rather than useless and inconsequential) was that for the all-reflector system. If any work needed to be done, it was to investigate the operation of the all-reflector system. The CPSC did not do that, although it made official recommendations that its requirements in the all-reflector system be obeyed beyond the CPSC's lawful scope of authority; that is, during use of the bicycle instead of merely while the bicycle is in interstate commerce.
In summary, the CPSC's reasoning about its requirement for the all-reflector system shows no familiarity with the concepts on which that system is advocated or criticized, no additional data concerning the effect of the system, and its study was not designed to discover anything useful about the system.
So far the discussion of the defects of the all-reflector system has concentrated on its deficiencies in providing protection from the front, where it attempts to substitute, dangerously, for the headlamp. The all-reflector system also reduces the effectiveness of the rear reflector, the one direction in which a reflector can provide a reasonable degree of protection. The all-reflector system is based on the idea (one that we know is false) that providing a reflector facing in every direction alerts all motorists that might collide with the bicycle. Because of the optical properties of reflectors, each reflector can operate only over a limited range of angles. The reflector's brightness diminishes somewhat as the angle from the centerline increases, until at some limiting angle the reflector completely extinguishes as the light illuminating the reflector goes completely through it instead of being reflected. This means that a large number of reflectors, about 12, is required to cover the complete azimuth circle. In order to avoid the problem of mounting reflectors in 12 different positions on a bicycle, the reflector industry developed what it called wide-angle reflectors. These are nothing more, in concept, than three reflectors, each facing in a different direction, that are molded in one piece. This enables the complete azimuth circle to be covered with four wide-angle reflectors, each covering 90 degrees of arc. The true invention is not the concept of three reflectors in one, it is in making the molds that enable this to be accomplished economically.
The problem with the wide-angle reflector is that it cannot be as bright as the standard automotive reflector. It may not be as bright per square inch of surface area, and in any case it has only one-third of its area pointing in any one direction. Therefore, any wide-angle reflector is, at most, only one-third as bright as a conventional reflector made with the same technology.
The CPSC made two other design decisions about rear reflectors. The first was to emphasize detection at extreme range, about 1,000 feet. The CPSC assumed that detection at greater range would markedly reduce the number of car-bike collisions. Even with a speed differential of 60 mph, detection at 1,000 feet gives more than 11 seconds of notice. This is far greater than is needed, and is far beyond the usual range of driver attention. The trouble with emphasizing detection at extreme range is that it reduces the brightness of the reflector at more useful ranges, the ranges at which collision avoidance action actually occurs. The second defective design decision was to insist on red color. The colors of reflectors are produced solely by absorbing those colors in the original light that are not desired. The reflector cannot produce any light by itself and it cannot change the color of any light that falls upon it. As a result of absorbing more of the illuminating light, red reflectors are only about one-third as bright as amber reflectors.
As a result, the requirements that were determined by these design decisions produced rear reflectors that are only about 10 percent as bright as reflectors of the same size, weight, and cost that are made for automobiles and other purposes, but are unlawful for sale for use on bicycles.
It is important to note that many pedals sold today cannot accept the CPSC-required pedal reflectors and are used without them. Bicycles equipped with these pedals are sold openly. Pedal reflectors that face forward cannot substitute for the headlamp any more than the front reflector can, because the problem is not one of position on the bicycle but the direction of the motorist's headlamp beams. However, pedal reflectors that face backward are very popular with motorists, who comment that these are often the only reflectors that they see. There are two reasons for this. First, on many bicycles the rear reflector has been removed or is covered by clothing or baggage. This is because of the CPSC's requirement that it be located where those items do obscure it. Second, the CPSC-required rear reflector is only about 10 percent as bright as it could be, so that the pedal reflectors appear much brighter in consequence. When compared against a proper rear reflector, the pedal reflectors do not appear brighter. Therefore, if a proper rear reflector were used, there is no advantage to retaining the pedal reflectors.
There is the argument that the motion of pedal reflectors attracts attention and therefore reduces collisions. There has been no empirical test of the this theory, and there is theoretical evidence against it. Motion has an attractive effect only when the movement appears in the peripheral vision. Motorists driving in daylight spend most of their time looking where they are going, and at night the effect is even stronger. If the motorist is looking where he is going, he will see a bright object that is moving slowly in his field of vision just about as easily as he will see one that is moving vigorously. The motion effect of pedal reflectors, although popular with the general public, probably has little effect in reducing car-bike collisions.
The front and side reflectors of the all-reflector system are frequently not illuminated by the headlamp beams of the car that collides with the cyclist.
These types of collision are the majority of car-bike collisions that are caused by darkness.
The all-reflector system must be completely useless when the other party does not project powerful headlamp beams ahead of it, as pedestrians and bicycles do not.
The majority of the population believes that the all-reflector system is a satisfactory substitute for the use of a headlamp.
The overwhelming majority of the nation's traffic experts, as expressed by the National Committee For Uniform Traffic Laws and Ordinances, rejects the all-reflector system because doing so would make motorists liable for hitting cyclists whom they could not see.
The all-reflector system produces rear reflectors, those facing in the only direction in which reflectors operate reliably, that are only about 10 percent as bright as other reflectors commonly available.
The all-reflector system must be done away with. There is no way that it can be made to work in the engineering sense of providing a reasonable degree of protection when cycling at night, and there is no reasonable way to prevent its presence from misleading the public and the industry in the continuing belief that the all-reflector system does provide adequate protection when cycling at night.
The CPSC must remove its requirement for the all-reflector system from its bicycle regulation, removing all of the requirements for that system and starting again from scientific and engineering principles that are supported by data. The CPSC must retract and correct its official finding that the all-reflector system "provides adequate visibility to motorists under lowlight conditions." The CPSC must issue a public statement that only a headlamp and a bright rear reflector can provide a reasonable degree of protection when cycling at night, although additional equipment is allowed.
The CPSC may, indeed should, create a standard for the optical performance of bicycle headlamps. This standard should specify the minimum intensity of light required over a solid angle that covers a horizontal arc of about 70 degrees on each side of the center line and a vertical arc of about 10 degrees. The amount of light should be calculated as that required to be visible at the distance given in the typical traffic laws of the states, which is 500 feet. The regulation must say that the requirements of this standard address the risks of car-bike, bike-pedestrian, and bike-bike collisions when cycling at night, as is specified in the state traffic laws.
Practically speaking, there is no need for a maximum limit, because of the difficulty of obtaining sufficient power to produce a dangerously bright light, but there would be no harm in providing an upper limit that is comparable to the upper limits allowed for the lower beams of the headlamps of motor vehicles.
There is no need to specify a particular ability to illuminate the cyclist's path, such as is in the requirements of several European nations. These have been unsatisfactory, producing ill-shaped beams that do not match the cyclist's needs. The required amount of illumination of the cyclist's path varies greatly with the actual current use. As we know, American urban areas are frequently so well lighted that cyclists can see their paths sufficiently well to travel as they desire. However, in other areas cyclists require much more illumination to be able to see where they are going. The required width of the beam also varies with use. Cycling on bikepaths generally requires a much wider beam than does cycling on the road, and off-road cycling requires a wider beam still. The required height of the beam also varies with speed. With these variations in requirement, it is impossible to produce a mandatory standard that would be satisfactory.
It is probably unnecessary to produce requirements for duration of lighting. At the present time, lights are available with durations of lighting varying from an hour or so to, theoretically, infinity for generators. Cyclists have not had a problem knowing the duration to be expected from any one light. It may be useful to require the manufacturer to state the duration to be expected from any particular set of batteries, a statement that many make at this time.
It is probably unnecessary to produce requirements for environmental resistance. Lights that have been used in the past have operated well under the conditions that cyclists will accept for cycling.
There is one desirable characteristic that has disappeared over the decades of the all-reflector system; that is, the standard lamp mounting bracket. In former times, most lamps were equipped with a standard socket that fitted over a matching standard bracket that was mounted on the bicycle. The typical bracket was about 1.25" square and about 0.1" thick. Each bicycle manufacturer could place the bracket wherever was most appropriate for the type of cycling the bicycle was intended to provide, and the manufacturers of lightweight bicycles equipped theirs with removable brackets that were held on with one bolt, so that all the cyclist had to carry for daylight rides was the one socket for the bolt.
It would be desirable for the CPSC to specify the dimensions of a standard lamp bracket, without requiring that all bicycles and all lamps have matching brackets. Then, lamp manufacturers would have the choice of either providing their own mounting system or of using the standard bracket. At least, all those who chose to use the bracket would produce interchangeable products.
It would also be desirable to specify the dimensions of a standard mounting system for generators. I have proposed that the standard be two 5-mm holes 30 mm apart, approximately in a vertical direction, in a plate 2 to 3 mm thick. What makes this system useful is that it uses an intermediate plate between the bracket that is brazed onto the bicycle's fork and the generator itself. The intermediate plate is made of soft aluminum, so that it may be easily bent to the appropriate angle and drilled to take the generator at the appropriate location to touch the tire properly. That way, it is easy to adapt to a different brand or style of generator or to a different size of tire. A full description of this system appears in my book Effective Cycling.
The CPSC should adopt as its rear reflector requirements those for the SAE style amber reflector, about 3" in diameter, that is typically available from auto parts stores. This has proved to be amply bright in most situations, and certainly is much brighter than the current CPSC reflector. It would be desirable that those intended for bicycle use be equipped with a standard threaded stud projecting from the back, of a size commonly used on bicycles, say 5 mm x 0.8, although a hole for such a bolt would be acceptable. The standard must say that the requirements of this standard address the risk of being hit from behind by a faster vehicle when cycling at night.
It is undesirable to specify the location of the reflector in the CPSC standard, because all possible locations have disadvantages under different using conditions. The existing state traffic laws handle the situation by requiring that the reflector, at night, be visible from behind; if the user loads his baggage to obscure the reflector and does not move it to a location where it can be seen from behind, then he is disobeying the law.
For the same reason, there is no reason to specify requirements for brackets on which to mount the rear reflector. If the reflector is specified to be mounted with a 5-mm bolt, then brackets will be made to accept that 5-mm bolt. How the other end of the bracket attaches to the bicycle then depends on the bicycle manufacturer and the user.
The question may be raised that this dramatic change in the CPSC's requirements for nighttime safety equipment will require changes in the traffic laws across the nation. Yes, it will, but that will not be more than a momentary problem. So long as the CPSC states that these changes address the risks of car-bike, bike-pedestrian and bike-bike collisions when cycling at night, as is described above, local laws are automatically changed. The requirements that are peculiar to one traffic code or another, such as the requirement for reflectors on the end of the pedal, the requirements that caused the bicycle industry so much grief that it advocated the Consumer Product Safety Commission Improvements Act, will be automatically invalidated. Every state requires a white headlamp, so those laws will remain in force. Every state requires a rear reflector, although most specify that it be red. The only change that will be required is from red to amber, and perhaps a restatement of distance.
Law enforcement would still retain its power to cite cyclists with no headlamp or no reflector before that change gets made. With the simple requirement that the headlamp and the rear reflector are the only equipment that is required when cycling at night, instead of fussing over the requirements for several different reflectors even in the daytime, law enforcement personnel ought to be favorably impressed and more able to develop the initiative to act to improve public behavior, either by prosecution or by the education that ought to precede prosecution.
The changes required by the bicycle industry are minor. The industry may stop providing on all bicycles the reflectors that it has been providing. If it provides rear reflectors on any model of bicycle it must change over to that required by the new CPSC standard, the SAE type that is currently available, probably at no difference in cost. Over time, different brackets for rear reflectors will evolve, each suited for a particular mounting location.
The industry may choose to provide the standard headlamp bracket with some bicycles, or not, as it thinks most suitable for the market. Component makers will design and produce brackets with standard heads to match the lamps, probably designed in a variety of styles to suit different mounting places on the bicycle. Some lamp makers will continue with their proprietary mounting systems, others will adopt the standard lamp socket to fit the standard bracket. In the same way some generator makers will retain their proprietary mounting systems while others, in conjunction with bicycle manufacturers, will adopt the standard bracket. Probably, both lamp brackets and reflector brackets will evolve to provide quick, simple and reliable mounting and dismounting of equipment, so that cyclists will use it when required while leaving it off when not required.
The bicycle industry will have to revise their bicycle safety booklets to say that a headlamp and a rear reflector are required when cycling at night to both obey the law and to provide a reasonable degree of protection from car-bike collisions and to other road users, such as pedestrians and other cyclists.
Following these recommendations is necessary to develop a cycling public that acts to become reasonably safe when cycling at night, although other actions, such as in education and law enforcement, are also necessary. The recommended requirements provide the equipment that has been shown necessary for reasonably safe cycling at night, both for the individual cyclist, the motorist who at least will suffer property damage in a collision with the cyclist, and for other non-motorized road users.
However, if these recommendations are not followed, cyclists will continue to be confused about the proper nighttime safety equipment and will continue to be killed and injured much more frequently than can be achieved under a rational law that is based on engineering knowledge. Then the task of education and law enforcement would remain as it is today, practically impossible because of the confusion produced by the dangerous and confusing present requirement for the all-reflector system.
1 Fed Reg 16 July 74, pg 26100
2 I have misplaced the particular page in the Federal Register that contains this finding.
3 Brezina, E. & M. Kramer; An Investigation of Rider, Bicycle and Environmental Variables in Urban Bicycle Collisions; Ontario Dept. of Transport; Oct 1970; found in CPSC record at pp 1063-1100
4 If the cyclist has the duty to yield but does not, the collision is not caused by the motorist's failure to observe the cyclist. This type of collision occurs frequently in daylight when visibility is not a consideration, and hence is not one of the prime nighttime accident types.
5 Kenneth Cross; Identifying Critical Behavior Leading to Collisions Between Bicycles and Motor Vehicles; Office of Traffic Safety, State of California; June, 1974
6 Forester vs CPSC, No. 75-1292, Court of Appeals for the District of Columbia, 1976-77
7 Cross, Kenneth D. & Gary Fisher; A Study of Bicycle/Motor-Vehicle Accidents: Identification of Problem Types and Countermeasure Approaches; National Highway Traffic Safety Administration, Sept. 1977
8 Forester, John; Bicycle Transportation pg 93; the MIT Press, 1983
9 Essex County Court of Common Pleas, New Jersey, 1993
10 Atkinson, J.E. and P.M. Hurst; Collisions between cyclists and motorists in New Zealand; Accid. Anal. Prev. 15:137-151;1983
11 Petty, Ross D.; Bicycle Safety: A Case Study i Regulatory Review; Regulation, 1994 No 2, pp22-24. Petty has written other papers on this subject.
Return to: John Forester's Home Page Up: Lights