At issue in this suit is the constitutionality of certain gun laws enacted by the District of Columbia. The district court determined as a matter of law that the District’s efforts “to combat gun violence and promote public safety” by means of its registration laws were “constitutionally permissible.” Heller v. District of Columbia, 45 F.Supp.3d 35, 38 (D.D.C.2014). Before this court, Dick Anthony Heller *269and his co-appellants challenge both the district court’s admission of, and its reb-anee upon, certain expert reports proffered by the District and the final order denying Heller’s and granting the District’s motion for summary judgment.
We hold the district court’s admission of the challenged expert reports was not an abuse of discretion. We affirm in part and reverse in part the district court’s judgment in favor of the District.
I. Background
In District of Columbia v. Heller (Heller I) the Supreme Court held the District of Columbia’s “prohibition of handguns held and used for self-defense in the home” was unconstitutional. 554 U.S. 570, 636, 128 S.Ct. 2783, 171 L.Ed.2d 637 (2008). Immediately thereafter, the D.C. City Council revised the District’s gun laws by enacting the Firearms Registration Amendment Act of 2008(FRA). D.C. Law 17-372.
The FRA created a “new scheme for regulating firearms.” Heller v. District of Columbia, 670 F.3d 1244, 1248 (D.C.Cir.2011) (Heller II). With limited exceptions, the FRA required the registration of all firearms in the District. D.C.Code § 7-2502.01. The law also imposed various conditions upon the registration of a firearm and limited the persons eligible to register a firearm by excluding, for example, individuals who within the prior five years had been convicted of certain drug or violent crimes or had a severe mental health problem, and individuals under the age of 18. Id. § 7-2502.03-.07. In addition, the FRA required the gun owner to renew the registration of his firearm(s) every three years, id. § 7-2502.07a, and prohibited registration — and hence possession — of certain firearms, such as short-barreled rifles and assault weapons. Id. § 7-2502.02.
• In July 2008 Heller filed suit challenging the District’s new registration scheme as inconsistent with the Second Amendment to the Constitution of the United States. The district court granted summary judgment to the District and Heller appealed.
On that appeal, we upheld the constitutionality of the District’s “basic registration requirement,” insofar as that requirement pertained to handguns. Heller II, 670 F.3d at 1254-55. We also upheld the portion of the FRA prohibiting registration, and therefore possession, of assault weapons and magazines with a capacity in excess of 10 rounds. Id. at 1247-48, 1264.
We reserved judgment as to the constitutionality of the District’s basic registration requirement for long guns, the conditions under which a registration certificate would be issued, and the duration for which such a certificate would be valid. Id. at 1255, 1258-60. We held that both the basic registration requirement for long guns, if not de minimis, and the conditions for registration were subject to intermediate scrutiny, and that the record as it then stood was not sufficient for us to evaluate whether those laws were narrowly tailored to serve an important governmental interest. Id. at 1258. We therefore remanded the case to the district court for further evidentiary proceedings. Id. at 1260.
Subsequently, the D.C. Council enacted the Firearms Amendment Act of 2012, D.C. Law 19-170, which repealed certain of the conditions for registration, such as the requirement that a pistol be submitted for ballistic identification as part of the registration process, and reduced the burden upon registrants imposed by other provisions. Heller then filed an amended complaint to take account of these legislative changes.
During discovery, Heller and the District offered the opinion testimony of, respectively, one and four expert witnesses. *270Heller v. District of Columbia, 45 F.Supp.3d at 40 (Heller III). Largely upon the basis of their testimony, the district court entered summary judgment for the District.
On this appeal, Heller argues the district court erred by admitting the opinion testimony of three of the District’s four expert witnesses. In addition, Heller argues the district court erred in upholding as constitutional: (1) the basic registration requirement as it pertains to long guns, D.C.Code § 7-2502.01(a); (2) the requirement that one appear in person to register a firearm and be fingerprinted and photographed, id. § 7-2502.04; (3) the requirement that the registrant bring with him the firearm to be registered, which requirement the Metropolitan Police Department (MPD) may or may not invoke as to a particular individual, id. § 7 — 2502.04(c); (4) the expiration of the registration after three years, id. § 7-2502.07a; (5) the imposition of certain fees for registration, id. § 7-2502.05(b); (6) the requirements that a registrant complete a firearms safety and training course or provide evidence of another form of training and that the registrant pass a test to demonstrate his knowledge of the District’s firearms laws, id. §§ 7-2502.03(a)(13), 7-2502.03(a)(10); and (7) the prohibition on registration of more than one pistol per person in any 30-day period, id. § 7-2502.03(e).
II. Analysis
We first address the district court’s admission of the challenged expert reports and related testimony. We then turn to Heller’s constitutional challenges.
A. The expert reports and testimony
Heller moved to strike three of the four expert reports offered by the District during discovery, viz., those of Cathy Lanier, the Chief of the MPD, and of Mark Jones and Joseph Vince, Jr., both former agents of the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF), but not that of Daniel Webster, Director of the Johns Hopkins Center for Gun Policy. Heller argued the expert reports “fall short of the disclosure requirements under Fed. R.CrvP. 26(a) and that their proposed testimony [was] too unreliable to be admitted under Fed.R.Evid. 702.” The district court denied Heller’s motion.
On appeal, Heller renews both arguments. We review the district court’s admission of expert testimony for abuse of discretion, whether that admission is challenged under the rules of evidence or under the rules of procedure. United States v. Day, 524 F.3d 1361, 1367 (D.C.Cir.2008). We conclude that the district court did not abuse its discretion in admitting the challenged testimony.
1. Federal Rule of Civil Procedure 26
Federal Rule of Civil Procedure 26(a)(2)(B) provides that an expert witness must submit a written report containing, among other things, “a complete statement of all opinions the witness will express and the basis and reasons for them,” as well as “the facts or data considered by the witness in forming them.” A party who fails to comply with Rule 26(a)(2)(B) generally may not use that witness “to supply evidence on a motion ... unless the failure was substantially justified or is harmless.” Fed.R.Civ.P. 37(c)(1).
The purpose of the rule is to avoid “unfair surprise to the opposing party.” Muldrow ex rel. Estate of Muldrow v. Re-Direct, Inc., 493 F.3d 160, 167 (D.C.Cir.2007) (citation and internal quotation marks omitted). Admitting a report with an omission that does not cause “unfair surprise” we deem harmless. Id.
*271As the district court noted, each of the challenged expert reports contained an explicit statement as to the basis for that witness’s opinion, to wit, that his or her report was based “on my experience, my review of numerous studies and books, the District of Columbia’s firearms laws and regulations, and discovery materials from this case made available to me.” In addition each report recounts in detail the expert’s relevant experience. The district court stated:
Plaintiffs have had an opportunity to depose these experts and examine more fully the bases for their opinions.... Where Defendants have provided adequate notice of the opinions they expect these experts to offer and Plaintiffs have had and continue to have opportunities to challenge these conclusions, the goals of Rule 26(a) are satisfied, and there is no basis for striking the reports and preventing these experts from testifying.
Heller v. District of Columbia, 952 F.Supp.2d 133, 139 (D.D.C.2013).
The district court did not abuse its discretion in so holding. The experts’ reports adequately established the bases for the opinions they expressed in the reports and in their declarations. Heller had the opportunity to probe the bases for the witnesses’ opinions when he deposed them.
2. Federal Rule of Evidence 702
Rule 702 provides:
A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if:
(a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;
(b) the testimony is based on sufficient facts or data;
(c) the testimony is the product of reliable principles and methods; and
(d) the expert has reliably applied the principles and methods to the facts of the case.
Fed.R.Evid. 702.
In Daubert v. Merrell Dow Pharmaceuticals, the Supreme Court held Rule 702 requires courts to ensure that expert testimony is “not only relevant, but reliable.” 509 U.S. 579, 589, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993); see also Kumho Tire Co. v. Carmichael, 526 U.S. 137, 149, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) (noting that “Daubert’s general principles apply” not just to scientific testimony but to all “the expert matters described in Rule 702”). Therefore, courts are obligated to “determine whether [expert] testimony has a reliable basis in the knowledge and experience of [the relevant] discipline.” Kumho Tire, 526 U.S. at 149, 119 S.Ct. 1167 (second alteration in original) (citation and internal quotation marks omitted). Nonetheless, the Supreme Court has said “the trial judge must have considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable.” Id. at 152, 119 S.Ct. 1167.
In this case the district court reasoned:
[I]t appears here that the opinion evidence is connected to the existing facts—the registration requirements and the state of gun violence in the District—by a methodology precisely contemplated by Daubert and Rule 702: each expert’s professional judgment obtained through long experience in the field. Each of the reports specifically identifies this experience as being the basis for the opinions proffered, and each provides some justification—in the form of information gained from the ex*272pert’s relevant experience — for those opinions.
Heller v. District of Columbia, 952 F.Supp.2d at 142.
As the district court rightly suggested, each of the challenged experts has decades of relevant experience. Still, the Advisory Committee notes to Rule 702 provide that a witness who is “relying solely or primarily on experience ... must explain how that experience leads to the conclusion reached, why that experience is a sufficient basis for the opinion, and how that experience is reliably applied to the facts.” In this case the experts’ explanation of the connection between their experience and their conclusions was sometimes fatally sparse. Likewise, the district court failed meaningfully to evaluate the factual bases for the experts’ opinions, noting only that they were supported by “some justification — in the form of information gained from the expert’s relevant experience.”
As this court has noted, however, the “admission of [expert] testimony does not constitute an abuse of discretion merely because the factual bases for an expert’s opinion are weak.” Joy v. Bell Helicopter Textron, Inc., 999 F.2d 549, 567 (D.C.Cir.1993). Nor is this a case in which the experts’ reports consisted of “subjective belief or unsupported speculation,” which the rules of evidence preclude. Id. at 570 (citation and internal quotation marks omitted).
In addition to invoking his or her generalized “experience,” each expert claimed to have relied upon specific news stories, academic studies, or other research in forming an opinion. Moreover, each of the three experts was in a position to state whether the cited materials comported with his or her personal experience.
In light of the challenged experts’ substantial relevant experience and the sources they cited in support of their conclusions — the above-noted stories, studies, and research — we hold the district court did not abuse its discretion in admitting the challenged expert reports and the subsequent expert declarations. Rather, as the district court noted, Heller’s “concerns about the conclusions [to which] these experts’ experience led them ... go to the weight of the testimony,” not its admissibility. Heller v. District of Columbia, 952 F.Supp.2d at 142.
B. The constitutional challenges
We review the district court’s summary judgment determination de novo, considering the evidence in the light most favorable to the non-moving party, ie., Heller. See Ayissi-Etoh v. Fannie Mae, 712 F.3d 572, 576 (D.C.Cir.2013).
In Heller II, we adopted a two-step approach to determining the constitutionality of the District’s gun registration laws: “We ask first whether a particular provision impinges upon a right protected by the Second Amendment; if it does, then we go on to determine whether the provision passes muster under the appropriate level of constitutional scrutiny.” 670 F.3d at 1252. We determined that level was intermediate scrutiny. Id. at 1252-53.
For a challenged provision to survive intermediate scrutiny, the District has to show, first, that it “promotes a substantial governmental interest that would be achieved less effectively absent the regulation,” and second, that “the means chosen are not substantially broader than necessary to achieve that interest.” Id. at 1258 (quoting Ward v. Rock Against Racism, 491 U.S. 781, 782-83, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989)). To meet the first requirement, the District must demonstrate that the harms to be prevented by the regulation “are real, not merely conjectural, and that the regulation will in *273fact alleviate these harms in a direct and material way.” Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622, 662-64, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994) (Turner I). We do not, however, review de novo the District’s evidence of the harm to be prevented and the likely efficacy of the regulation in preventing that harm. See id. at 666, 114 S.Ct. 2445. Rather, it is our remit to determine only whether the District “has drawn reasonable inferences based on substantial evidence.” Id. If it has done so, and if the means chosen are not overbroad, then “summary judgment ... is appropriate regardless of whether the evidence is in conflict.” Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 185, 195-96, 211, 117 S.Ct. 1174, 137 L.Ed.2d 369 (1997) (Turner II); see also Heller II, 670 F.3d at 1263 (upholding the District’s ban on assault weapons on the basis that “the evidence demonstrates a ban on assault weapons is likely to promote the Government’s interest in crime control”).
1. Impingement
In Heller II we held the basic registration requirement as applied to handguns did not impinge upon the Second Amendment and was therefore constitutional. 670 F.3d at 1254-55 (“[T]he basic requirement to register a handgun is longstanding in American law.... Therefore, we presume the District’s basic registration requirement including the submission of certain information does not impinge upon the right protected by the Second Amendment. Further, we find no basis in either the historical record or the record of this case to rebut that presumption.”) (citations omitted); see also Heller I, 554 U.S. at 626-27 & n. 26, 128 S.Ct. 2783 (“longstanding” firearm regulations are “presumptively lawful”). We left open the question whether requiring the registration of long guns impinges upon the Second Amendment. 670 F.3d at 1255 n.* *; see also D.C.Code § 7-2502.01(a). We now hold it does not.
Requiring the registration of handguns is legally different from requiring the registration of long guns only in that “basic registration of handguns is deeply enough rooted in our history to support the presumption that [it] is constitutional,” Heller II, 670 F.3d at 1253; the registration requirement for long guns lacks that historical pedigree. Id. at 1255.
Even absent the presumption that attends the pedigree, however, the basic registration requirement as applied to hand guns falls into the category of requirements that are “self-evidently de minimis, for they are similar to other common registration or licensing schemes, such as those for voting or for driving a car, that cannot reasonably be considered onerous.” Id. at 1254-55. On Heller’s previous appeal, we were unable to determine whether requiring the registration of long guns is similarly a de minimis burden because the record was “devoid of information concerning the application of registration requirements to long guns.” Id. at 1255 n.* *. We therefore allowed Heller, during the discovery proceedings on remand, the opportunity to introduce evidence that might differentiate the registration requirement for long guns from other registration requirements that undoubtedly entail a de minimis burden upon a constitutional right. As the district court subsequently determined, however, Heller offered no evidence distinguishing the basic registration requirement as applied to long guns. See Heller III, 45 F.Supp.3d at 51. Indeed, he did not even argue the point.1
*274Because the burden of the basic registration requirement as applied to long guns is de minimis, it does not implicate the second amendment right. Heller II, 670 F.3d at 1255; see also Justice v. Town of Cicero, 577 F.3d 768, 773-75 (7th Cir.2009) (holding local ordinance “requiring the registration of all firearms” is consistent with the Supreme Court’s ruling in Heller I). It is therefore constitutional.
The additional registration requirements, however, cannot be said to be de minimis. In Heller II, we held the additional requirements, as they then stood, “affect[ed] the Second Amendment right because they [we]re not de minimis” — that is, they “ma[d]e it considerably more difficult for a person lawfully to acquire and keep a firearm ... for the purpose of self-defense in the home.” Id. at 1255. The subsequent repeal of some of those requirements and the amendment of others somewhat reduced the burden imposed upon District residents’ exercise of their Second Amendment rights. The District does not go so far as to argue, however, that the amended requirements are de minimis. Those requirements are therefore subject to intermediate scrutiny.
2. Intermediate scrutiny
We previously identified two substantial governmental interests served by the registration requirements enacted by the District: (1) protecting police officers by enabling them to determine, in advance, whether guns may be present at a location to which they are called and (2) aiding in crime control. Heller II, 670 F.3d at 1258. On remand, the District recharacterized the second interest as a broader interest in “promoting public safety.” Heller III, 45 F.Supp.3d at 49. On appeal, the District identifies more particularly its interest in “protecting police officers” and reiterates its interest in “promoting public safety” generally.
Heller does not dispute that these are substantial governmental interests. Rather, he challenges the closeness of the fit between the asserted interests and the various registration requirements. We agree with Heller that the District has not offered substantial evidence from which one could draw a reasonable conclusion that the challenged requirements will protect police officers; but we think the District has pointed to substantial evidence that some of the requirements — but not others — will promote public safety.
a. Police protection
Heller argues the registration requirements do not advance the District’s interest in protecting the police because MPD officers very rarely check the regis*275tration records in responding to a call, conducting an investigation, or executing a search warrant. The District responds that although the “MPD does not routinely check registration records prior to responding to a call for service ... such a check is a tool available for use in appropriate circumstances.” It is undisputed that such checks have taken place, albeit rarely.
Therefore, the question remains whether that “tool” promotes the District’s asserted interest in police protection. Discovery subsequent to our decision in Heller II indicates it does not.
According to the deposition testimony of an MPD officer, District police “are trained to treat situations where there might be a crime in progress or domestic dispute or some other situation possibly involving violence as always having a potential to have a dangerous weapon present.” Further, one of the District’s expert witnesses stated that if the registration system indicated no weapon was present at an address, then officers “would continue to exercise caution.” The best the District’s expert could offer was that positive confirmation of a gun might raise officers’ “caution level ... that much higher.”
The testimony of the District’s own witnesses, therefore, indicates that the records established via the registration requirements, when queried at all, have little to no effect upon the conduct or safety of police officers. In light of this additional evidence, we agree with the statement of our colleague in Heller II that the asserted interest in police protection “leaves far too many false negatives to satisfy .... intermediate scrutiny.” 670 F.3d at 1295 (Ka-vanaugh, J., dissenting).
b. Public safety
Drawing directly upon the Report of the Judiciary Committee of the D.C. Council with respect to the Firearms Amendment Act of 2012, the District claims the various registration requirements advance its interest in public safety by “distinguishing criminals from law-abiding citizens, enabling police to arrest criminals immediately, facilitating enforcement against prohibited persons obtaining or continuing to possess firearms, reducing gun trafficking, and increasing the difficulty for criminals to acquire guns.” We next address whether the District has, with regard to each challenged registration provision, offered substantial evidence from which it could reasonably have concluded the provision will mitigate various threats to public safety “in a direct and material way,” Turner I, 512 U.S. at 664, 114 S.Ct. 2445, whether in one of the ways anticipated by the D.C. Council or otherwise.
i. In-person appearance, fingerprinting, and photographing, D.C .Code § 7-2502.04
The District has presented substantial evidence from which it could conclude that fingerprinting and photographing each person registering a gun promotes public safety by facilitating identification of a gun’s owner, both at the time of registration and upon any subsequent police check of the gun’s registration. The requirement that registrants appear in person is necessary in order for a photograph and fingerprints to be taken.
First, the fingerprinting requirement: The Report of the Committee on the Judiciary stated that “[t]he initial fingerprinting requirement is fundamental for the [MPD] to fulfill its public safety obligations in registering firearms — being able to screen the registrant to ensure that he or she is not disqualified from possessing a firearm.” In support of this assertion, the District points to the testimony of Chief *276Lanier, who said “[u]sing biometrics [ie., fingerprints] to positively identify an individual is far more effective than relying simply on a name and social security number.” Chief Lanier reiterates this conclusion in her expert declaration, and it is echoed in Webster’s expert declaration.
In addition, the District points to evidence suggesting background checks using fingerprints are more reliable than background checks conducted without fingerprints, which are more susceptible to fraud. Specifically, the District points to an ’ investigation conducted by the U.S. Government Accountability Office, in which five “agents acting in an undercover capacity used ... counterfeit driver’s licenses in attempts to purchase firearms from gun stores and pawnshops that were licensed by the federal government to sell firearms.” GAO-01-427, FIREARMS Purchased from Federal Firearm Lioensees Using Bogus Identifioation 2 (2001). Those attempts were, without exception, successful. Id. at 2-3. The report concluded that federal background checks conducted by the firearm dealers “cannot ensure that the prospective purchaser is not a felon or other prohibited person whose receipt and possession of a firearm would be unlawful.” Id. at 2.2
Heller argues the District has not experienced a problem with fraud in the registration of firearms. He also implies the problem is unlikely to arise, given the increased difficulty of manufacturing fraudulent identification documents today, as compared to 2001, when the GAO concluded its investigation. Even if this is true, however, a prophylactic disclosure measure such as the one at issue here survives intermediate scrutiny if the deterrent value of the measure will materially further an important governmental interest. See Barry v. City of New York, 712 F.2d 1554, 1559-61 (2d Cir.1983) (upholding under intermediate scrutiny a law requiring financial disclosures by certain publicly employed individuals in the face of a right-to-privacy challenge on the basis that it could “help deter corruption,” despite a “virtually corruption-free history” (citation and internal quotation marks omitted)). The GAO study indicates the fingerprinting requirement would, indeed, help to deter and detect fraud and thereby prevent disqualified individuals from registering firearms.
Regarding the requirement of a photograph: The Committee on the Judiciary emphasized “the importance of a registrant being able to present a registration certificate with a photograph, so police can quickly identify whether and to whom the firearm has been legally registered.” The Committee pointed to the testimony of Chief Lanier, who asserted that “a certificate with a photo helps to quickly and safely communicate” the fact of registration to police officers, which, “in turn, helps to keep both the officer and the registrant safe.” Heller, while maintaining that photographing a registrant will not deter fraud, does not contest that photographic confirmation of a registrant’s identity would be beneficial to public safety when the police encounter an armed registrant. See D.C.Code § 7-2502.08(c) (“Each registrant shall have in the registrant’s possession, whenever in possession of a firearm, the registration certificate, or exact photocopy thereof, for such firearm, *277and exhibit the same upon the demand of a member of the [MPD], or other law enforcement officer”).
For the foregoing reasons, we believe the District has adduced substantial evidence from which it reasonably could conclude that fingerprinting and photographing registrants will directly and materially advance public safety by preventing at least some ineligible individuals from obtaining weapons and, more important, by facilitating identification of the owner of a registered firearm during any subsequent encounter with the police. Those requirements are therefore not unconstitutional. The additional requirement that registrants appear in person to be photographed and fingerprinted is but a corollary necessary to implement those requirements. See Heller II, 670 F.3d at 1249 n. * (noting that administrative provisions “incidental to the underlying regime” are “lawful insofar as the underlying regime is lawful”).
ii. Bringing the firearm, D.C.Code § 7-2502.04(c)
The District argues that the “requirement that the firearm be made available for inspection allows MPD to verify that the application information is correct and that the firearm has not been altered or switched with another firearm.”' The District, however, has offered no evidence — -let alone substantial evidence— from which it can be inferred that verification will promote public safety. The district court acknowledged as much when it noted that not one of the District’s four experts “specifically addresse[d] the requirement that registrants bring the gun to be registered with them.” Heller III, 45 F.Supp.3d at 59. The district court nonetheless deemed it a “common-sense inference” that “if in-person appearance is necessary to verify the identity of the registrant, then physically bringing the gun is similarly necessary to verify the character of the registered weapon.” Id. Yet common sense suggests a person would not go to the trouble of obtaining a registration certificate for a weapon other than a weapon in his possession. On the contrary, common sense suggests that bringing firearms to the MPD would more likely be a threat to public safety; as Heller maintains, there is a “risk that the gun may be stolen en route or that the [would-be registrant] may be arrested or even shot by a police officer seeing a ‘man with a gun’ (or a gun case).”
iii. Re-registration, D.C.Code § 7-2502.07a
The District has offered three justifications for the requirement that a gun owner re-register his firearm every three years. None is supported by substantial evidence from which the District could reasonably have concluded that requiring reregistration would advance an important governmental interest.
First, the District’s experts argued that re-registration “will improve public safety by making sure that, in the time since [the gun owner] first registered, [he has] not fallen into a category of persons prohibited from owning a firearm.” Heller III, 45 F.Supp.3d at 67-68. As Heller rightly points out, however, “District officials and experts conceded [that] background checks could be conducted at any time without causing the registrations to expire.” The re-registration requirement cannot survive intermediate scrutiny on the (dubious) basis that it will make this task easier. Cf. McCullen v. Coakley, — U.S. -, 134 S.Ct. 2518, 2540, 189 L.Ed.2d 502 (2014) (“To meet the requirement of narrow tailoring, the government must demonstrate that alternative measures that burden substantially less speech would fail to achieve *278the government’s interests, not simply that the chosen route is easier”).
Second, the District argues triennial reregistration will help to maintain the accuracy of the registration database. This seems self-evidently true, but it is far from an adequate reason for burdening every gun owner when there is already a requirement that gun owners report relevant changes in their information, such as a new address. D.C.Code § 7-2502.08 (requiring such reporting). To the extent that a gun owner’s death or disposal of a registered gun is a fact of which the District should be aware, the District’s registration requirements as applied to any new owner within the District should satisfy that interest.
Third, the District argues that it has “an interest in its residents verifying the whereabouts of their firearms” in order “to determine when firearms have been lost or stolen.” District law, however, separately requires a gun owner to report the loss or theft of a weapon “immediately upon discovery” of the loss or theft, and imposes a monetary penalty for failure to do so. Id. § 7 — 2502.08(a)(1). In contrast, the re-registration provision imposes no penalty for failure to re-register except the revocation of one’s registration certificate, but a person whose weapon has been lost or stolen no longer has need of a certificate. Although the District fails to make the argument express in its brief, the report of its Committee on the Judiciary, on which the brief relies in general, asserted that the re-registration provision may complement the loss-reporting provision because it “likely causes the owner to look for his or her gun if it hasn’t been used” for a while, but that is mere speculation. The re-registration process requires only that a gun owner affirm that he still has the registered weapon; it does not require the gun owner physically to examine the weapon. See id. § 7-2502.07a. Therefore, there is no reason to believe that an owner who does not suspect his gun has been lost or stolen is likely to look for the registered weapon prior to re-registering it.
iv. Fees, D.C.Code § 7-2502.05
Heller argues “[t]he District may not condition exercise of a fundamental constitutional right on the creation of a burdensome registration regime and then justify imposing ‘administrative costs’ to pay for it.” He does not argue the registration fees of $13 per firearm and $35 for fingerprinting, D.C. Mun. Regs. tit. 24, § 2320.3(c)(3), are unreasonably high.
As we already said in Heller II, “administrative ... provisions incidental to the underlying regime” — which include reasonable fees associated with registration — are lawful insofar as the underlying regime is lawful. 670 F.3d at 1249 n. *; see also Cox v. New Hampshire, 312 U.S. 569, 577, 61 S.Ct. 762, 85 L.Ed. 1049 (1941) (holding, in response to a First Amendment challenge to a parade licensing statute, that a government may impose a fee “to meet the expense incident to the administration of the act and to the maintenance of public order in the matter licensed”); Kwong v. Bloomberg, 723 F.3d 160, 165-69 (2d Cir.2013) (holding constitutional a $340 fee for a license to possess a handgun in one’s home). As such, reasonable fees associated with the constitutional requirements of registration and fingerprinting are also constitutional.
v. Education requirements, D.C.Code §§ 7-2502.03(a)(10), 7-2502.03(a)(13)
The District has presented substantial evidence from which it could conclude that training in the safe use of firearms promotes public safety by reducing accidents involving firearms, but has presented no evidence from which it could *279conclude that passing a test of knowledge about local gun laws does so. The safety training, therefore, is constitutional; the test of legal knowledge is not.
Regarding the one-hour firearms safety course, available online or at the MPD, FiReaems Safety Training Course, https:// dcfst.mpdconline.com/ (last visited Aug. 21, 2015), the District’s experts each testified to their belief in the value of training to prevent accidents. Heller responds that “the District’s experts cite no studies showing that mandatory training or testing in gun safety reduce unintentional discharges.” The District, however, need not present such evidence. Rather, the Supreme Court has “permitted litigants to justify ... restrictions ... based ... on history, consensus, and simple common sense” when the three are conjoined. Cf. Lorillard Tobacco Co., 533 U.S. 525, 555, 121 S.Ct. 2404 (2001) (internal quotation marks omitted). In this case, the District has offered anecdotal evidence showing the adoption of training requirements “in most every law enforcement profession that requires the carrying of a firearm” and a professional consensus in favor of safety training.3 Though its experts have characterized the training requirement as a matter of “common sense,” this is not a case in which the District has asked the court to rule upon the basis of “common sense” alone.
None of the District’s experts, however, offers any reason to believe that knowledge of the District’s gun laws will promote public safety. Indeed, the closest the District’s experts came to addressing the subject was the statement by Chief Lanier that “in order to make registrants more clearly accountable under the law, it is important to be able to demonstrate that they were taught and aware of the requirements.” This assertion, however, does not tie knowledge of the law to the District’s interest in public safety.
Furthermore, even if acquiring knowledge of the law were demonstrably helpful, the imposition of a requirement that registrants prove their knowledge of the law on “a test prescribed by the Chief’ is an additional burden, see D.C.Code § 7-2502.03(10), the utility of which is supported by no evidence whatsoever, not even anecdotal evidence. Moreover, only a few of the 15 questions in the test actually prescribed by the Chief plausibly reflect a concern with public safety.4
Because the District has offered no evidence from which the court can infer it reasonably concluded that knowledge of its gun laws, as shown by passing its test, will promote public safety, on this record the requirement must be held constitutionally invalid.
vi. One-pistol-per-month rule, D.C.Code. § 7-2502.03(e)
The District has not presented substantial evidence to support the conclu*280sion that its prohibition on the registration of “more than one pistol per registrant during any 30-day period,” D.C.Code § 7-2502.03(e), “promotes a substantial governmental interest that would be achieved less effectively absent the regulation.” Heller II, 670 F.3d at 1258 (quoting Rock Against Racism, 491 U.S. at 782-83, 109 S.Ct. 2746). It is therefore unconstitutional.
The District argues that the limitation could reduce gun trafficking and that it would further promote public safety by limiting the number of guns in circulation, as the District “could reasonably conclude that more guns lead to more gun theft, more gun accidents, more gun suicides, and more gun crimes.”
As for the District’s first argument, what little expert testimony it presented indeed indicates that limiting gun purchases in turn might limit trafficking in weapons. The experts’ conclusion that limiting gun registrations would likewise reduce trafficking is, however, unsupported by the evidence. For example, Chief Lanier stated “[s]tudies have shown that laws restricting the registration or purchase of multiple firearms in a given period are effective in disrupting illegal interstate trafficking of firearms.” Yet the only study she and the District’s other witnesses cited has nothing to do with “laws restricting registration,” as its title attests. See Douglas S. Weil & Rebecca C. Knox, Effects of Limiting Handgun Purchases on Interstate Transfer of Firearms, 275 J. Am. Med. Ass’n 1759 (1996). One of the experts also testified from his own observation that when Virginia limited firearm purchases to one every 30 days, fewer guns bought in Virginia were used in crimes committed in the District; traffickers, he observed, instead sourced more guns through straw purchasers in Maryland. But even if this is true, the suggestion that a gun trafficker would bring fewer guns into the District because he could not register more than one per month there lacks the support of experience and of common sense. Indeed, as Heller notes, even Chief Lanier acknowledged that the efficacy of purchasing limitations in preventing trafficking may have little bearing upon the efficacy of registration limitations in doing so.
As for the District’s second argument, one of its experts testified that, in his opinion, “the most effective method of limiting misuse of firearms, including homicide, suicide, and accidental injuries, is to limit the number of firearms present in a home.” Accepting that as true, however, it does not justify restricting an individual’s undoubted constitutional right to keep arms (plural) in his or her home, whether for self-defense or hunting or just collecting, because, taken to its logical conclusion, that reasoning would justify a total ban on firearms kept in the home. See Parker v. District of Columbia, 478 F.3d 370, 400 (D.C.Cir.2007), aff'd sub nom. Heller I (rejecting the District’s argument that a ban on one type of gun was constitutional because the “prohibition ... [did] not threaten total disarmament” and noting that, if such argument were adopted “[i]t could similarly be contended that all firearms may be banned so long as sabers were permitted”).
III. Conclusion
For the reasons set forth above, the district court’s final order is AFFIRMED with respect to: the basic registration requirement as applied to long guns, D.C.Code § 7-2502.01(a); 'the requirement that a registrant be fingerprinted and photographed and make a personal appearance to register a firearm, D.C.Code § 7-2502.04; the requirement that an individual pay certain fees associated with the registration of a firearm, D.C.Code § 7-*2812502.05; and the requirement that registrants complete a firearms safety and training course, D.C.Code § 7-2502.03(a)(13). The district court’s order is REVERSED with respect to the requirement that a person bring with him the firearm to be registered, D.C.Code § Y — 2502.04(c); the requirement that a gun owner re-register his firearm every three years, D.C.Code § 7-2502.07a; the requirement that conditions registration of a firearm upon passing a test of knowledge of the District’s firearms laws, D.C.Code § 7-2502.03(a)(10); and the prohibition on registration of “more than one pistol per registrant during any 30-day period,” D.C.Code § 7-2502.03(e).
So ordered.
. In his reply brief in this court, Heller argued for the first time that the registration requirement impinges upon the Second Amendment right to bear arms because a *274person can "go to prison and receive a lifetime ban on possession of firearms for failure to register or reregister.” See D.C.Code §§ 7-2502.03, 7-2507.06, 7-2502.08 (providing generally violation of the registration requirements may result in fines, imprisonment, and ineligibility to register weapons in the future). This assertion, however, is too little, too late. It comes too late because we do not ordinarily notice an argument that first appears in a reply brief. See Gunpowder Riverkeeper v. FERC, No. 14-1062, -F.3d-, -, 2015 WL 4450952, at *5 (D.C.Cir. July 21, 2015) ("[AJrguments not clearly raised in a party's opening brief are generally considered to be forfeit”). In any event, it is too little because in Heller II we instanced other licensing schemes we think impose a de minimis burden notwithstanding that failure to comply with those schemes may result in criminal penalties; so it is with the basic registration requirement for long guns. See Heller II, 670 F.3d at 1254-55 (describing licensing schemes "such as [that] for ... driving a car” as "self-evidently de minimis”); D.C.Code § 50-1403.01(e) (providing that an individual found guilty of "operating a motor vehicle in the District” while that person’s license is "revoked or suspended” may be fined or imprisoned for up to one year).
. The states in. which the GAO conducted its study had adopted the National Instant Criminal Background Check System (NICS), see 18 U.S.C. § 922(t), under which, then as now, the following information is required of each individual who undergoes a NICS check: (1) name, (2) sex, (3) race, (4) date of birth, and (5) state of residence. 28 C.F.R. § 25.7. A dealer may, in addition, report the purchaser’s Social Security or other identifying number and physical description. Id.
. J.A. 394 (Lanier declaration) ("California, Connecticut, Hawaii, Massachusetts, and Michigan all have laws requiring some sort of training or safety certification as part of the registration process, and other jurisdictions are considering instituting similar requirements"); J.A. 407 (Vince declaration) (stating that he “do[es] not know of one firearm expert or law enforcement trainer who has not strongly recommended attending and successfully passing a safety course prior to owning or using a firearm”).
. Compare J.A. 834 ("When handling a firearm, you should always: (A) Treat it as if it is loaded; (B) Point it in a safe direction; (C) Both A and B”) with J.A. 834 ("To purchase ammunition in the District of Columbia you must have the following in your possession: (A) A U.S. Passport; (B) A valid firearm registration certificate; (C) A valid driver's license”).