United States Court of Appeals
                            FOR THE EIGHTH CIRCUIT
                                   ___________

                                   No. 00-3457
                                   ___________

United States of America,               *
                                        *
            Appellee,                   *
                                        * Appeal from the United States
      v.                                * District Court for the
                                        * Northern District of Iowa
Wade Allen Wheat,                       *
                                        *
            Appellant.                  *
                                   ___________

                             Submitted: April 10, 2001
                             Filed: December 28, 2001 (Corrected: 01/04/02)
                                    ___________

Before LOKEN, Circuit Judge, BOGUE,1 District Judge, and GOLDBERG,2
Judge.3
                               ___________

GOLDBERG, Judge.



      1
          The Honorable Andrew W. Bogue, United States District Judge for the
District of South Dakota, sitting by designation.
      2
         The Honorable Richard W. Goldberg, Judge of the United States Court of
International Trade, sitting by designation.
      3
         Pursuant to 28 U.S.C. § 46(b), the Chief Judge certified the existence of a
judicial emergency necessitating the designation of a panel consisting of fewer than
two members of the Court of Appeals.
      Wade Allen Wheat appeals from his conviction on one count of possession
of cocaine base (“crack”). He argues, inter alia, that the district court4 erred in
denying his motion to suppress, and that, even if his conviction were lawful, his
sentence is unconstitutional in light of the Supreme Court’s decision in Apprendi
v. New Jersey, 

(2000). Because we find that the District Court
properly denied Wheat’s motion to suppress and that the Apprendi error was
harmless, we affirm.

I.    FACTUAL AND PROCEDURAL BACKGROUND

       On May 3, 1996, a motorist used his cellular phone to place a 9-1-1 call to
the Blairsburg, Iowa Police Department. The caller reported that a tan- and cream-
colored Nissan Stanza or “something like that,” whose license plate began with the
letters W-O-C, was being driven erratically in the northbound lane of Highway
169, eight miles south of Fort Dodge, Iowa. The caller complained that the Nissan
was passing on the wrong side of the road, cutting off other cars, and otherwise
being driven as if by a “complete maniac.” The 9-1-1 operator did not ask the
caller to identify himself.

       Police dispatchers relayed the caller’s tip to patrolling officers. Shortly
thereafter, Officer Paul Samuelson observed a tan Nissan Maxima whose license
plate began with the letters W-O-C, stopped in the northbound lane of Highway
169 at the intersection of Highway 20. The Nissan made a right turn, and Officer
Samuelson stopped it immediately, without having observed any incidents of
erratic driving. He obtained the driver’s licenses of the driver and Wheat, who
was sitting in the front passenger seat; there were no other passengers. A check
disclosed that Wheat’s license was suspended but that notice of the suspension had


      4
         The Honorable Mark W. Bennett, Chief Judge, United States District Court
for the Northern District of Iowa.

                                        -2-
never been served. Because Officer Samuelson was unfamiliar with the procedure
for service, he radioed Officer Aaron Anderson, whom he already had an
appointment to meet, to request assistance at the scene.

       Shortly after Officer Anderson arrived, a dispatcher radioed Officer
Samuelson to inform him that the suspension had in fact already been served.
However, because Officer Samuelson had noticed that the driver’s hands were
fidgeting, he asked Officer Anderson whether he had any previous experience with
the driver. Officer Anderson told Officer Samuelson that the driver had a history
of drug problems and that he had run from the police on several occasions. On the
basis of this information, and after informing Wheat that the suspension did not
need to be served, Officer Samuelson requested and received permission from the
driver to search the vehicle. At Officer Samuelson’s request, the driver exited the
vehicle.

       As Officer Samuelson was walking around toward the passenger side,
Wheat opened his door and exited the vehicle on his own initiative. When Officer
Samuelson reached the passenger side, he noticed a dry brown paper bag from a
McDonald’s restaurant at Wheat’s feet. Because it was raining during the duration
of the stop, the dryness of the bag was remarkable. Suspecting that Wheat had just
discarded the bag, Officer Samuelson retrieved it, and found that it held four
smaller plastic bags containing what appeared to be a controlled substance. A
further search of the vehicle also revealed a small quantity of marijuana. Around
fifteen minutes after he first pulled them over, Officer Samuelson arrested both the
driver and Wheat for possession of controlled substances. Subsequent laboratory
testing proved the contents of the plastic bags to be 63.03 grams of crack cocaine.

       Wheat was indicted by a grand jury on one count of possession with intent
to distribute more than 50 grams of a mixture or substance containing cocaine
base, in violation of 21 U.S.C. § 841(b)(1)(A)(iii) (1994). He pled not guilty and

                                        -3-
filed a motion to suppress all evidence obtained and statements made at the May 3,
1996 vehicle stop. The motion was denied.

       In a two-day jury trial beginning March 10, 1997, the appellant was found
not guilty of possession with intent to distribute, but was convicted on the lesser
included offense of simple possession of cocaine base, pursuant to 21 U.S.C. §
844(a). The district court sentenced the appellant to 110 months imprisonment.
After several procedural twists and turns not relevant to the issues before us now,
this appeal followed.

II.   DISCUSSION

      Wheat’s principal contentions are that the district court erred by denying his
motion to suppress all evidence obtained as a result of the May 3, 1996 stop of the
vehicle in which he was a passenger, and, failing the success of that argument, that
his 110-month sentence is unconstitutional in light of the Supreme Court’s
subsequent decision in Apprendi v. New Jersey, 

(2000).

A.    The District Court Properly Denied the Motion to Suppress

       We conduct a de novo review of the district court’s denial of the motion to
suppress. Ornelas v. United States, 

, 699 (1996). However, we
review the findings of historical fact that underlie the district court’s decision for
clear error, and afford “due weight” to the inferences drawn from those facts by
the district court and the participating law enforcement officers. Id.; see also
United States v. Ball, 

, 262 (8th Cir. 1996).

      Wheat argues that the anonymous 9-1-1 call could not give rise to
reasonable suspicion sufficient to justify an investigatory stop under Terry v.
Ohio, 

(1968), because Officer Samuelson never witnessed any traffic

                                          -4-
violation actually in progress or about to occur. Wheat also argues that he and the
driver of the Nissan should have been allowed to leave the scene immediately after
Officer Samuelson discovered that no license suspension needed to be served.5

      1.     Officer Samuelson Had Reasonable Suspicion to Initiate the Stop

       When a law enforcement officer directs a motor vehicle to stop by the side
of the road and detains its occupants for questioning, such an investigatory stop
constitutes a search and seizure under the Fourth and Fourteenth Amendments,
“even though the purpose of the stop is limited and the resulting detention quite
brief.” Delaware v. Prouse, 

, 653 (1979); accord Whren v. United
States, 

, 809-10 (1996); see also Thomas v. Dickel, 

,
1024 (8th Cir. 2000). Under Terry and its progeny, “[a]n investigatory stop is
permissible under the Fourth Amendment if supported by reasonable suspicion.”
Ornelas v. United States, 

, 693 (1996); see also United States v.
Sharpe, 

, 682 (1985) (applying Terry to investigatory stop of
vehicle); United States v. Bell, 

, 749 (8th Cir. 1999) (“An
investigative stop does not violate the Fourth Amendment if the police have
reasonable suspicion that the vehicle or its occupants are involved in criminal
activity.”). If the investigatory stop is not justified by reasonable suspicion or if
the investigating officers exceed the stop’s proper scope, any evidence derived
from the stop is inadmissible at trial. See, e.g., Wong Sun v. United States, 

, 484 (1963); United States v. Ramos, 

, 1164 (8th Cir. 1994).
A passenger in a motor vehicle has standing to challenge the stop of that vehicle.
See United States v. Lyton, 

, 1170 (8th Cir. 1998).

      5
        Before this Court, Wheat does not suggest, as he did before the district court,
that the search of the paper McDonald’s bag was unlawful. As the government
nonetheless takes the trouble to argue in its brief to the Court, such an argument
would fail, as Wheat had no legitimate expectation of privacy in the surface of the
pavement near his car. United States v. Dawdy, 

, 1430 (8th Cir. 1995).

                                         -5-
       Because reasonable suspicion is a less demanding standard than the
probable cause required for an arrest, it “can arise from information that is less
reliable than that required to show probable cause,” including an anonymous tip.
Alabama v. White, 

, 330 (1990). Whether an anonymous tip suffices
to give rise to reasonable suspicion depends on both the quantity of information it
conveys as well as the quality, or degree of reliability, of that information, viewed
under the totality of the circumstances. 

tip has a relatively low degree
of reliability, more information will be required to establish the requisite quantum
of suspicion than would be required if the tip were more reliable.” 

the Supreme Court considered whether the requisite quantum of
suspicion was established by an anonymous tip claiming that a named individual
would leave a specific apartment at a particular time and transport an ounce of
cocaine in a brown Plymouth station wagon with a broken tail light to a specific
motel. 

The Court stated that the tip itself provided virtually no
indication that the caller was honest or that the tip was reliable, as it gave no basis
for its predictions. 

However the Court held that visual corroboration
by law enforcement officers of most aspects of the tip, including the suspect’s sex,
the time of her departure, the vehicle she drove, and her apparent destination, gave
the tip sufficient indicia of reliability. 

(“When significant aspects of the
caller’s predictions were verified, there was reason to believe not only that the
caller was honest but also that he was well informed, at least well enough to justify
the stop.”).

        Subsequent to White, with respect to two categories of anonymous tips,
lower courts tended to find reasonable suspicion even where such tips lacked
personal corroboration by law enforcement officers of any predictive elements.
First, several federal appellate courts held that verification by police of the
innocent details of a tip, such as the suspect’s description, could amount to
reasonable suspicion when the principal allegation of the tip was that the suspect

                                         -6-
was armed with a gun and presented a potentially immediate danger. See, e.g.,
United States v. Clipper, 

, 946-51 (D.C. Cir. 1992); United States v.
Bold, 

, 102-04 (2d Cir. 1994); United States v. DeBerry, 

,
885-87 (7th Cir. 1996); United States v. Gibson, 

, 619-25 (11th Cir.
1995); see also United States v. Roberson, 

, 81 n.4 (3d Cir. 1996)
(speculating in dicta that “a different rule may apply” if an anonymous tip alleged
that the suspect possessed a weapon rather than drugs). The courts agreed that the
suspect’s alleged possession of a firearm was an important factor to be considered
in weighing the totality of the circumstances, because the “element of imminent
danger distinguishes a gun tip from one involving possession of drugs.” 

; accord 

(“Armed persons are so
dangerous to the peace of the community that the police should not be forbidden
to follow up a tip that a person is armed, and as a realistic matter this will require a
stop in all cases.”); 

; 

. The courts
recognized that “an officer who corroborates every item of information reported
by an anonymous tipster other than actual possession is left with an unappealing
choice. He must either stop and frisk the individual, or wait to see if he ultimately
brandishes or uses the firearm.” 

(internal quotation marks
and citations omitted) (citing 

).

       The second category of tips for which courts declined to require
corroboration of their predictive elements concerned a different type of potentially
immediate threat, and the one at issue in this case: an apparently drunk or reckless
driver. Thus, a number of state supreme and intermediate appellate courts6 held
that law enforcement officers could pull over a vehicle for an investigatory stop
based on a contemporaneous tip of erratic driving that accurately described a


      6
         Presumably because traffic violations are adjudicated in state rather than
federal court, there is a paucity of reported federal court decisions concerning Fourth
Amendment challenges to Terry stops based on anonymous tips of erratic driving.

                                          -7-
given vehicle, even where the officer did not personally witness any moving
violations and therefore lacked probable cause to make an arrest. See, e.g., State
v. Melanson, 

, 200-03, 

, 339-41 (1995); State v.
Sampson, 

, 1327 (Me. 1996); State v. Lamb, 

, 196-203,

, 1102-06 (1998); State v. Slater, 

, 696-706, 

, 1041-46 (1999); see also State v. Markus, 

(Iowa Ct. App.
1991); State v. Smith, 

(Ind. Ct. App. 1994); People v. Rance, 

, 

(N.Y. App. Div. 1996); Kaysville City v. Mulcahy,

(Utah Ct. App. 1997). But see McChesney v. State, 

,
1075-78 (Wyo. 1999) (3-2 decision) (holding that corroboration only of color,
make, and direction of suspect vehicle anonymously reported to be weaving and
dangerously passing other cars did not give rise to reasonable suspicion where
officer did not personally witness erratic driving after trailing vehicle a substantial
distance); State v. Miller, 

, 640-45 (N.D. 1994) (holding that tip
that driver in fast-food restaurant’s drive-up lane “could barely hold his head up”
and was possibly drunk could not create reasonable suspicion because it was
“short on reliability, . . . short on specifics,” and uncorroborated by police officer);
State v. Lee, 

, 393-96, 

, 638-40 (1997) (holding that
anonymous caller’s “belief” that driver was under influence of alcohol and was
speeding, without any indication that the caller’s belief was based on personal
observation of drinking, speeding, or erratic driving, did not justify investigatory
stop); see also State v. Villegas-Varela, 

, 114-19, 

,
810-13 (1994) (holding that anonymous tip that did not give location or direction
of suspect car that was ultimately not stopped until more than an hour later did not
give rise to reasonable suspicion). In addition to the fact that in the erratic driving
context the tipster is almost invariably claiming to describe contemporaneously
perceived behavior, see, e.g., 

, 665 A.2d at 340, courts
tended to agree that, as in the gun possession cases, the exigency of the situation
demanded an immediate law enforcement response. See, e.g., 

665 A.2d
at 340.

                                          -8-
       Last year, however, the Supreme Court appeared to curtail the argument that
a purported threat of imminent danger necessarily lessens the government’s
burden in an analysis of the reliability of an anonymous tip. In Florida v. J.L., 

(2000), a decision not cited by the parties to the instant case, the Court
considered whether an anonymous tip that a person is carrying a gun is sufficient
to justify a Terry stop and frisk. An anonymous caller had reported that a young
black male wearing a plaid shirt and standing at a particular bus stop was carrying
a gun; the caller did not identify himself or explain whence he knew such
information. 

271. Police officers responding to the call spotted a
young black male in a plaid shirt at the designated bus stop and, though they saw
no firearm and had no reason to suspect the individual or his companions of illegal
conduct, nevertheless stopped and frisked him, seizing a gun. 

       A unanimous Court held that the stop-and-frisk was unconstitutional, as
“the bare report of an unknown, unaccountable informant who neither explained
how he knew about the gun nor supplied any basis for believing he had inside
information about” the suspect lacked even the “moderate indicia of reliability
present in White.” 

The Court expressly rejected both the argument that
the officers’ confirmation of the suspect’s visual attributes provided sufficient
corroboration of the tip, 

and the argument that an “automatic firearm
exception” should apply to the standard Terry analysis. 

While
recognizing the danger posed by firearms, the Court expressed its fears that an
automatic exception would facilitate harassment based on false tips, and that a
similar exception would be claimed for “bare-boned tips about narcotics,” since
those who possess large quantities of drugs are frequently assumed to be armed.
See 

At the same time, the Court declined

      to speculate about the circumstances under which the danger alleged
      in an anonymous tip might be so great as to justify a search even
      without a showing of reliability. We do not say, for example, that a
      report of a person carrying a bomb need bear the indicia of reliability

                                         -9-
      we demand for a person carrying a firearm before the police can
      constitutionally conduct a frisk. Nor do we hold that public safety
      officers in quarters where the reasonable expectation of Fourth
      Amendment privacy is diminished, such as airports and schools,
      cannot conduct protective searches on the basis of information
      insufficient to justify searches elsewhere.

(citations omitted).

       The question we now face is whether, in light of J.L., an anonymous tip
about the dangerous operation of a vehicle whose innocent details are accurately
described may still possess sufficient indicia of reliability to justify an
investigatory stop by a law enforcement officer who does not personally observe
any erratic driving. Recognizing the complexity of this issue, we answer
affirmatively, and hold that under the totality of the circumstances of this case,
Officer Samuelson had reasonable suspicion to detain the car in which Wheat was
a passenger.

       In reaching this conclusion, we have been influenced by the reasoning
employed by those state courts that have already considered the issue. The
Supreme Courts of Vermont, Iowa, and Wisconsin have held that J.L. does not
prevent an anonymous tip concerning erratic driving from acquiring sufficient
indicia of reliability to justify a Terry stop, even when the investigating officer is
unable to corroborate that the driver is operating the vehicle recklessly and
therefore unlawfully.

       In the first such decision by a state high court, State v. Boyea, 

(Vt. 2000), cert. denied, 

(2001), the Supreme Court of Vermont
affirmed by a 3-2 vote the denial of a motion to suppress where the investigating
officer’s sole initial basis for detaining the motorist was an anonymous tip, relayed
by a police dispatcher, describing a “blue-purple Volkswagen Jetta with New York


                                         -10-
plates, traveling south on I-89 in between Exits 10 and 11, operating erratically.”

868. Although the court admitted that the case was close, see 

n.7, it distinguished J.L. on several grounds. First, it stated that the informant’s
accurate description of the vehicle and correct prediction of its location, just
minutes before the stop, gave the tip greater reliability than the bare-bones tip in
J.L. Second, the court observed that “[i]n contrast to the report of an individual in
possession of a gun, an anonymous report of an erratic or drunk driver on the
highway presents a qualitatively different level of danger, and concomitantly
greater urgency for prompt action.” 

The court noted that whereas
police confronted with a report of a concealed gun could quietly observe the
suspect “for a reasonable period of time without running the risk of death or injury
with every passing moment[, a]n officer in pursuit of a reportedly drunk driver on
a freeway does not enjoy such a luxury.” 

a driver is not unlike the
“bomb” for which, as the Supreme Court suggested in J.L., a laxer standard of
reliability may apply. 

court reasoned that the liberty interest
implicated by a simple motor vehicle stop was weaker than the “hands-on
violation of the person” that occurred in J.L.. 

       A very similar analysis informed the decision of the Supreme Court of Iowa
in State v. Walshire, 

(Iowa 2001). The court found that because
a tip about erratic driving describes not concealed criminal activity, as in J.L., but
rather “illegality open to public observation,” it therefore “demonstrated the
tipster’s basis of knowledge” and its reliability could be demonstrated through
corroboration of innocent details. 

Unlike the unknown informant in
J.L., who “neither explained how he knew about the gun nor supplied any basis for
believing he had inside 

, the caller in Walshire made
it clear that he or she was a private citizen eyewitness to an ongoing 

. The court also agreed with the Boyea court that the imminent
danger of drunk driving “might call for a relaxed threshold of reliability,” and that



                                        -11-
the intrusion on privacy interests is slight in a vehicle stop as compared to a pat-
down on a public street. 

       The Supreme Court of Wisconsin has also upheld an investigatory stop of a
vehicle based on an anonymous tip alleging erratic driving, albeit under somewhat
stronger factual circumstances. In State v. Rutzinski, 

, 

(2001), the court considered the reliability of the tip provided by an
unidentified motorist calling from a cell phone, contemporaneously describing a
black pickup truck that was weaving within its lane, variously driving too fast or
too slow, and tailgating. 

623 N.W.2d at 519. Although the
investigating officer did not personally observe any such erratic driving, 

623 N.W.2d at 519, the court affirmed the denial of the motion to suppress, on
three grounds. First, the court found that although the caller was never identified,
he or she did indicate during the phone tip that he or she was in the car directly in
front of the offending vehicle, and would therefore have been aware of the
possibility that the police could have traced his or her identity by recording his or
her license plate number. 

623 N.W.2d at 525. Next, the court
recognized that the caller explained that “he or she was making personal
observations of the [suspect driver’s] contemporaneous actions,” unlike the caller
in J.L.. 

623 N.W.2d at 526. Finally, like the Boyea court, the court
acknowledged that the tip in question suggested that the suspect “posed an
imminent threat to the public’s safety.” 

623 N.W.2d at 526. The
court emphasized that it did not advocate a blanket exception to the reliability
requirement for tips concerning alleged drunk driving, but at the same time clearly
stated that the “extraordinary danger” of drunk driving meant that any allegation
thereof had to be seriously considered in weighing the totality of the
circumstances. 

623 N.W.2d at 527.

      A handful of lower state courts to have considered this issue in light of J.L.
have reached a different conclusion, however. In Commonwealth v. Lubiejewski,

                                         -12-
49 Mass App. Ct. 212, 

(2000), a Massachusetts appeals court held
that an anonymous cellular phone call claiming that a pickup truck had been
driving on the wrong side of the road could not justify an investigatory stop. 

729 N.E.2d at 290-93. Although the trooper had corroborated all the
innocent details about the truck, including its license plate number, he had not
personally witnessed any erratic driving, and the caller had already told the
dispatcher that the truck had returned to the correct side of the road, so “the
emergency had ended.” 

729 N.E.2d at 292. In State v. Boyle, 

(La. Ct. App. 2001), a divided Louisiana appeals court held that an
anonymous tip that the driver of a particular pick-up truck was intoxicated did not
justify a Terry stop of the suspect in his own driveway. 

The court
focused on the fact that the officers had not witnessed any criminal activity or
unusual driving, and on the fact that the stop took place on the suspect’s private
property. 

In Washington v. State, 

(Ind. Ct. App.
2001), an Indiana appeals court reversed the denial of a motion to suppress
marijuana seized during an investigatory stop of a vehicle made pursuant to
anonymous tip that the driver was possibly drunk, because the officer had not
personally observed any dangerous driving during the two miles he followed the
car, and had not corroborated any detailed predictions of the suspect’s future
behavior. 

And in Stewart v. State, 

(Tex. Ct.
App. 2000), a Texas appeals court held that an anonymous tip that a driver had
appeared highly intoxicated while entering the vehicle did not justify a stop,
because the description of the suspect was so general that the court questioned
whether the officer was certain he had stopped the correct suspect, and the officer
had not witnessed any erratic driving. 

However, in State v. Marks,

(Conn. Super. Ct. Dec. 7, 2000), a Connecticut trial court
denied a motion to suppress a stop based on an anonymous tip of erratic driving,
even though the officer witnessed no unlawful driving and the tipster refused to
identify himself, because the tip took the form of a five-minute 9-1-1 call wherein
the caller “clearly relay[ed] his first hand observations as events [we]re unfolding

                                        -13-
before him,” describing in extensive detail “every erratic move” of the suspect’s
car. 

       From these cases, and upon further reflection, we believe the following
considerations to be integral to a determination of whether an anonymous tip of
erratic driving may justify an investigatory stop. In so doing, we bear in mind the
Supreme Court’s instruction that “[r]easonable suspicion, like probable cause, is
dependent upon both the content of information possessed by police and its degree
of reliability. Both factors--quantity and quality--are considered in the ‘totality of
the circumstances.’” 

(quoting United States v. Cortez, 

, 417 (1981)).

       First, the anonymous tipster must provide a sufficient quantity of
information, such as the make and model of the vehicle, its license plate numbers,
its location and bearing, and similar innocent details, so that the officer, and the
court, may be certain that the vehicle stopped is the same as the one identified by
the caller. The time interval between receipt of the tip and location of the suspect
vehicle, though going principally to the question of reliability, may also be a factor
here. Although the J.L. Court focused on deficiencies in the quality, rather than in
the quantity, of the information contained in the tip at issue in that case, we think
it significant that that tip only spoke of a young black male wearing a plaid shirt,
standing at a particular bus stop. See 

. That is a rather
generic description, and the possibility for confusion of the suspect’s identity was
compounded by the fact that the police only responded “sometime” later; the
record did not disclose how long. 

in the instant case, the caller
identified the color and make of the vehicle, named the first three letters of its
license plate, and gave its location and direction.7 Compare 

         Although the caller misidentified the model of the vehicle as a Stanza, rather
than as a Maxima, he expressed reservation, saying that it was “something like that.”
In all other respects his identification of the vehicle was correct.

                                         -14-
863 (caller accurately described a “blue-purple Volkswagen Jetta with New York
plates, traveling south on I-89 in between exits 10 and 11"), with 

(“[G]iven the generality of the radioed description, it is not clear
that the officer could even be sure that the automobile was being driven by the
man seen to [appear drunk] by the informer.”). See also 

n.18
(“Th[e] demand for specificity in the information upon which police action is
predicated is the central teaching of this Court’s Fourth Amendment
jurisprudence.”). Officer Samuelson effected a stop within minutes of the 9-1-1
call, and at the suppression hearing, he testified that he “knew right away that that
was the vehicle.” Tr. of Hr’g on Mot. to Suppress, at 20. We think that the
information in the tip identifying the recklessly driven vehicle was sufficiently
copious and precise.

      The tip must also contain a sufficient quantity of information to support an
inference that the tipster has witnessed an actual traffic violation that compels an
immediate stop.8 A law enforcement officer’s mere hunch does not amount to
reasonable suspicion, see 

; Illinois v. Wardlow, 

,
123-24 (2000); a fortiori, neither does a private citizen’s. Cf. Miller, 

         The rationale for allowing less rigorous corroboration of tips alleging erratic
driving is that the imminent danger present in this context is substantially greater (and
more difficult to thwart by less intrusive means) than the danger posed by a person
in possession of a concealed handgun. Therefore, the moving violation or violations
alleged must suggest real exigency. An allegation of erratic driving will generally
pass this test since it strongly suggests that the driver is operating under the influence
of alcohol or drugs and is unable to control his vehicle. So too would an anonymous
tip of drag racing or a game of “chicken”; at the other end of the spectrum, a report
of a vehicle being driven one mile per hour over the posted limit would almost
certainly not. But we need not delimit the offenses that might give rise to reasonable
suspicion, and indeed we cannot, for the inquiry must always be undertaken on a
case-by-case basis under the totality of the circumstances. In all cases, however, the
more extensive the description of the alleged offense, the greater the likelihood that
the tip will give rise to reasonable suspicion.

                                          -15-
at 640-45 (holding that tip that driver in fast-food restaurant’s drive-up lane “could
barely hold his head up” and was possibly drunk could not create reasonable
suspicion in part because it was “short on specifics”); 

,
938 P.2d at 638-40 (holding that anonymous caller’s “belief” that driver was under
influence of alcohol and was speeding, without any indication that the caller’s
belief was based on personal observation of drinking, speeding, or erratic driving,
did not justify investigatory stop). In the instant case, the anonymous caller
specifically alleged that he had personally observed several different traffic
violations involving erratic driving. See Iowa Code §§ 321.299 et seq. (2001)
(regulating proper methods of passing and overtaking another vehicle); Iowa Code
§ 321.277 (2001) (making reckless driving a simple misdemeanor).

       The second and far more difficult consideration concerns the quality, or
degree of reliability, of the information conveyed in an anonymous tip. In J.L., the
Court found the anonymous tip unreliable because it “provided no predictive
information and therefore left the police without means to test the informant’s
knowledge or 

. In the context of an anonymous tip of
erratic driving, however, the first challenge is to identify the predictive element of
the tip that must be corroborated. Is it that the vehicle is being driven
dangerously? Wholly aside from the public safety considerations discussed infra,
this seems too stringent, because a police officer who corroborated the claim of
reckless driving would then have not merely reasonable suspicion to justify an
investigatory stop, but probable cause to make an arrest. See United States v. Bell,

, 822 (8th Cir. 1996) (“[A]ny traffic violation, even a minor one, gives
an officer probable cause to stop the violator.”); accord United States v. Neumann,

, 755-56 (8th Cir. 1999); see also Iowa Code § 321.277. Thus, police
would lose the intermediate step of investigatory stops based on reasonable
suspicion. We think that under the proper circumstances Terry stops strike a
careful balance between society’s interest in crime prevention and the individual



                                        -16-
citizen’s interest in remaining free from wanton government interference, and we
are loath to see their use foreclosed in this context.9

       Alternatively, the predictive information to be corroborated might be no
more than that a vehicle matching a certain description will pass a certain
waypoint on a specific road at an approximate time. However, the Supreme Court
precluded this approach in J.L., at least with respect to gun possession cases.
While acknowledging that corroboration of the suspect’s “readily observable
location and appearance” does assure that the police have stopped the person that
the tipster meant to accuse, the Court stated that “[s]uch a tip . . . does not show
that the tipster has knowledge of concealed criminal activity. The reasonable
suspicion here at issue requires that a tip be reliable in its assertion of illegality,
not just in its tendency to identify a determinate person.” 

.10


      9
         Under Iowa law, a peace officer who has probable cause to believe a traffic
violation has been committed may, at her discretion, make a full-blown custodial
arrest of the suspect. See Iowa Code § 321.485(1)(a) (2001). A constitutional
scheme that absolutely forbade peace officers from immediately stopping a vehicle
reported to be dangerous, in order to ask a few questions of the driver, but encouraged
them to watch that same vehicle like hawks for the slightest violation of any of the
manifold motor vehicle regulations, and thereupon permitted them to make a full
custodial arrest, would do no service to individual liberties. Officers would have
incentive to issue citations or make arrests for very minor moving violations they
might otherwise overlook, in order to have some pretext to get a reportedly drunk
driver off the road. To do so would be perfectly constitutional, see Whren v. United
States, 

, 813 (1996), yet not something that we think the courts should
encourage.
      10
          Curiously, however, in White, which formed a cornerstone of the analysis
in J.L., none of the predictive information corroborated proved that the tipster had
knowledge of concealed criminal activity, a fact not lost on the dissent or on other
commentators. See 

(Stevens, J., dissenting) (“An anonymous
neighbor’s prediction about somebody’s time of departure and probable destination
is anything but a reliable basis for assuming that the commuter is in possession of an

                                         -17-
       A careful reading of the Supreme Court’s Fourth Amendment jurisprudence
suggests that this emphasis on the predictive aspects of an anonymous tip may be
less applicable to tips purporting to describe contemporaneous, readily observable
criminal actions, as in the case of erratic driving witnessed by another motorist.
White did not create a rule requiring that a tip predict future action, United States
v. Johnson, 

, 1125 (8th Cir. 1995), and neither did J.L.. As we have
previously acknowledged, “[s]uch a rule would be contrary to the line of cases
holding that reasonable suspicion must be judged on the totality of the
circumstances.” 

n.3. The Supreme Court has long emphasized that a
primary determinant of a tipster’s reliability is the basis of his knowledge. See
Illinois v. Gates, 

, 230 (1983) (“[A]n informant’s ‘veracity,’
‘reliability,’ and ‘basis of knowledge’ are all highly relevant in determining the
value of his report. . . . [T]hese elements should [not] be understood as entirely
separate and independent requirements to be rigidly exacted in every case . . . .
Rather, . . . they should be understood simply as closely intertwined issues.”);
accord 

. Unlike with clandestine crimes such as
possessory offenses, including those involving drugs or guns, where corroboration
of the predictive elements of a tip may be the only means of ascertaining the
informant’s basis of knowledge, in erratic driving cases the basis of the tipster’s
knowledge is likely to be apparent. Almost always, it comes from his eyewitness
observations, and there is no need to verify that he possesses inside information.
As the court reasoned in Boyea:

      The offense alleged here did not involve a concealed crime--a
      possessory offense. What was described in the police dispatch to the


illegal substance.”); 4 La Fave, Search and Seizure § 9.4(h), at 224-25 (3d ed. 1996)
(criticizing the majority opinion for crediting “‘the caller’s ability to predict
respondent’s future behavior,’ even when . . . that behavior--as far as the surveilling
officers can tell--is both routine and nonsuspicious”) (emphasis in original; footnotes
deleted).

                                         -18-
      arresting officer was a crime in progress, carried out in public,
      identifiable and observable by anyone in sight of its commission.
      Unlike the tip alleged in White--that White was carrying narcotics--
      . . . here a total stranger could have observed defendant’s driving
      abilities. No intimate or confidential relationship was required to
      support the accuracy of the observation. The caller simply reported a
      contemporaneous observation of criminal activity taking place in his
      line of 

(Skoglund, J., concurring) (emphasis in original). See also

(“[T]he information provided here did not
concern concealed criminal activity, but rather illegality open to public
observation. The tip here demonstrated the tipster’s basis of knowledge: the caller
observed the defendant driving in an erratic manner.”); Rutzinski, 241 Wis. 2d at

(“Unlike the caller in J.L., the informant in this case
provided the police with verifiable information indicating his or her basis of
knowledge. The informant explained that he or she was making personal
observations of Rutzinski’s contemporaneous actions.”). Cf. United States v.
Armstead, 

, 322 n.3 (8th Cir. 1997) (rejecting contention that
anonymity of airline agent made his or her tip about a drug courier unreliable, as
“[t]he exact identity of the airline’s agent is unimportant: it is enough that she or
he worked for the airline, . . . and would therefore have accurate information”).

       By way of contrast, in White corroboration of the predictive aspects of the
tip was the only means of confirming the tipster’s basis of knowledge. 

(“[T]he caller’s ability to predict respondent’s future behavior . . .
demonstrated inside information . . . [that t]he general public would have had no
way of knowing . . . . [A] person with access to [information about an individual’s
itinerary] is likely to also have access to reliable information about that
individual’s illegal activities.”) (emphasis in original). And in J.L., the tip lacked
any predictive elements that would suggest the tipster was in a special position to
know about the suspect’s clandestine 

(“All the

                                        -19-
police had to go on in this case was the bare report of an unknown, unaccountable
informant who neither explained how he knew about the gun nor supplied any
basis for believing he had inside information.”). Although the tipster did provide
an ostensibly contemporaneous account of the suspect’s actions, inasmuch as he
stated that the suspect was standing at a certain bus stop, he neither alleged that he
was observing any crime nor even explained how he knew that the suspect carried
a gun. See 

(“[The tip in question] does not show that the tipster has
knowledge of concealed criminal activity. The reasonable suspicion here at issue
requires that a tip be reliable in its assertion of illegality . . . .”). We think that an
anonymous tip conveying a contemporaneous observation of criminal activity
whose innocent details are corroborated is at least as credible as the one in White,
where future criminal activity was predicted, but only innocent details were
corroborated.11

       We recognize the danger that, as with any anonymous tip, even a
supposedly contemporaneous account of erratic driving could be a complete work
of fiction, created by some malicious prankster to cause trouble for another
motorist. Indeed, in J.L. the Supreme Court declined to adopt an automatic
firearm exception to the corroboration requirement for this very reason. See 529

      11
          Significantly, in our common law contemporaneous accounts of situations
have long been regarded as especially credible, and thus exempt from the hearsay rule
as present sense impressions, one aspect of the concept of res gestae. See Jon R.
Waltz, The Present Sense Impression Exception to the Rule Against Hearsay: Origins
and Attributes, 

. Rev. 869, 870-75 (1981) (explaining that this principle is
not, as often assumed, of recent origin, but rather dates back centuries). This
exception is of course now codified at Fed. R. Evid. 803(1) (excluding from the
hearsay rule “[a] statement describing or explaining an event or condition made while
the declarant was perceiving the event or condition,” regardless whether the declarant
is available as a witness). Although the declarant must have been a percipient
witness, i.e., “one who was in a position to perceive the event or condition which his
declaration purports to describe,” he need not be identified so long as there is no
doubt that a declaration was actually made. Waltz, 

. Rev. at 877-78.

                                           -20-
U.S. at 272. However, with respect to anonymous reports of erratic driving that
seem otherwise credible under the totality of the circumstances, we think that the
risk of false tips is slight compared to the risk of not allowing the police
immediately to conduct an investigatory stop, for several reasons.

       First, the risk that law enforcement officers themselves will fabricate such a
tip in order to harass innocent motorists is negligible. Where, as in this case, the
tip originates in the form of a 9-1-1 call, and is subsequently broadcast over the
police radio channel, there is no chance that the investigating officer has invented
the tip, a fear expressed in other circumstances by some commentators. See, e.g.,
4 La Fave, Search and Seizure § 9.4(h), at 227 (criticizing the Supreme Court’s
decision in Adams v. Williams, 

(1972), where the Court upheld an
investigatory stop based on a tip allegedly provided by informant known to officer
but otherwise never identified, on the grounds that it facilitates police
fabrications). That leaves the possibility of malicious hoaxes perpetrated by
private citizens. We wholeheartedly endorse efforts such as those mentioned by
Justice Kennedy in his concurrence in J.L., 

, to encourage law
enforcement to use instant caller identification technology or otherwise to try to
identify anonymous tipsters, in order to increase the reliability of such tips. In the
instant case, however, we confront a situation where the 9-1-1 operator failed to
ask the caller for his identity, and the caller did not volunteer such information.

       We must therefore decide whether that oversight, and the consequent
possibility that the allegation of erratic driving was groundless, destroyed the
reliability of the tip in question. From Terry, we know that

      [i]n order to assess the reasonableness of Officer [Samuelson’s]
      conduct as a general proposition, it is necessary first to focus upon
      the governmental interest which allegedly justifies official intrusion
      upon the constitutionally protected interests of the private citizen, for
      there is no ready test for determining reasonableness other than by

                                         -21-
      balancing the need to search [or seize] against the invasion which the
      search [or seizure] entails.

(initial brackets added; subsequent brackets in original;
internal quotation marks deleted). In cases of possible drunk driving, we think
that the governmental interest in effecting an immediate investigatory stop is very
strong, and the intrusion upon the constitutionally protected interests of the private
citizen, although also significant, is comparatively less so.

        An erratic and possibly drunk driver poses an imminent threat to public
safety. See 

-51, 623 N.W.2d at 526-27. Of course,
arguably so too does a citizen armed with a gun, yet the Supreme Court firmly
declined to adopt an automatic firearm exception to the reliability requirement on
that basis. 

. However, there is a critical distinction between
gun possession cases and potential drunk driving cases. In the possessory offense
cases, law enforcement officers have two less invasive options not available to
officers responding to a tip about a drunk driver. First, they may initiate a simple
consensual encounter, for which no articulable suspicion is required. See Florida
v. Rodriguez, 

, 5-7 (1984) (per curiam). Needless to say, that is not
possible when the suspect is driving a moving vehicle.12

       Alternatively, officers responding to a tip about a possessory violation may
quietly observe the suspect for a considerable length of time, watching for other
indications of incipient criminality that would give them reasonable suspicion to
make an investigatory stop--as, for example, in Terry, where an experienced
officer witnessed several men casing a 

. By contrast, where
an anonymous tip alleges erratic and possibly drunk driving, a responding officer


      12
           We do not suggest that the impracticality of conducting a consensual
interview obviates the requirement of reasonable suspicion, but we do think it is a
factor to be considered in weighing the government’s interest.

                                        -22-
faces a stark choice. As the Boyea court noted, he can intercept the vehicle
immediately and ascertain whether its driver is operating under the influence of
drugs or 

. Or he can follow and observe, with three
possible outcomes: the suspect drives without incident for several miles; the
suspect drifts harmlessly onto the shoulder, providing corroboration of the tip and
probable cause for an arrest; or the suspect veers into oncoming traffic, or fails to
stop at a light, or otherwise causes a sudden and potentially devastating accident.

to J.L., where the suspect was merely standing at the bus
stop, in this context the suspect is extremely mobile, and potentially highly
dangerous. See 

(“[A] drunk driver is not at all unlike a
‘bomb,’ and a mobile one at that.”). Thus, we think that there is a substantial
government interest in effecting a stop as quickly as possible.13

       That interest must be balanced against the individual’s right to remain free
from unreasonable government intrusion. See 

. We
readily acknowledge that citizens have a liberty interest in proceeding unmolested
along public highways and that investigatory stops create a substantial intrusion
on that interest. See Delaware v. Prouse, 

, 657 (1979) (noting that
vehicular stops interfere with freedom of movement, are inconvenient, consume
time, and may create substantial anxiety). At the same time, we think that such
stops are considerably less invasive, both physically and psychologically, than the
frisk on a public corner that was at issue in J.L.. See 

(reasoning that the liberty interest implicated by a vehicular stop is weaker than

      13
          Consequently, when the officer does not effect an immediate stop of a
potentially drunk driver, the force of this justification rapidly diminishes. See, e.g.,

(finding no reasonable suspicion where officer did
not personally witness erratic driving after trailing vehicle a substantial distance);

(Terry stop not justified where police caught up with
suspect in his own driveway); Washington v. 

(finding
no reasonable suspicion where officer had not personally observed any dangerous
driving during the two miles he followed the car).

                                         -23-
the “hands-on violation of the person” of the public frisk in J.L.); 

. Recognizing the “magnitude of the drunken driving problem,” the
Supreme Court has approved the use by law enforcement of brief roadside sobriety
checkpoints. See Michigan Dep’t of State Police v. Sitz, 

, 451
(1990). We rather doubt that a like recognition of the dangers of illegal handguns
would lead the Court to approve a scheme of randomly placed magnetometers
(“metal detectors”) on public sidewalks, even in high-crime neighborhoods.

       After careful consideration of all of the above factors, we find that the initial
stop of the vehicle in which Wheat was a passenger was not unreasonable under
the totality of the circumstances. An anonymous caller provided an extensive
description of a vehicle that, based on his contemporaneous eyewitness
observations, he believed was being operated dangerously, and cited specific
examples of moving violations. When Officer Samuelson caught up with the
vehicle minutes later while it was stopped at an intersection, he corroborated all its
innocent details, confirming that it was the one identified by the tipster. Within
seconds after the vehicle resumed motion, Officer Samuelson effected an
immediate investigatory stop, rather than allow it to proceed and potentially
endanger other vehicles. Under the totality of the circumstances, he had
reasonable suspicion to do so, and the stop was valid under the Fourth and
Fourteenth Amendments.

      2.     On Plain Error Review, Wheat Cannot Show That He Was
             Unreasonably Detained After Officer Samuelson Discovered That the
             Suspension Did Not Need to Be Served

      Wheat also claims that “after Samuelson discovered that no DOT
suspension needed to be served, Wheat should have been allowed to leave the
scene immediately.” Wheat’s Brief, at 16. However, Wheat failed to make this
argument before the district court. See infra. Reviewing it now for plain error, see


                                         -24-
Fed. R. Crim. Pro. 52(b); Johnson v. United States, 

, 465-66 (1997),
we find it to be without merit. As the district court discussed in a footnote:

      Defendant Wheat does not contend that he was unreasonably detained
      after Officer Samuelson learned that he did not have to serve a license
      suspension on him. Nor would such an argument have any merit.
      Immediately after learning that he did not have to serve the license
      suspension on Wheat, Officer Samuelson and Officer Anderson left
      Samuelson’s patrol car, returned Wheat’s license[,] and informed him
      that he did not need to be served with the license suspension.
      Immediately after informing Wheat of these events, Officer
      Samuelson turned to McDonald and requested consent to search the
      Nissan.

Gov’t App., at 51, Order Adopting Report and Recommendation, at 19 n.7
(emphasis in original).

       We must accept these findings of historical fact except upon a showing that
they are clearly erroneous, Ornelas v. United States, 

, 699 (1996), and
Wheat has offered no evidence to that end. Moreover, we agree with the district
court’s legal conclusions. “After making a valid Terry stop, police officers must
diligently work to confirm or dispel their suspicions in a short period of time.”
United States v. Bell, 

, 749 (8th Cir. 1999). In this case, Officer
Samuelson worked diligently to ascertain whether or not the license suspension
needed to be served; after learning that it did not, he immediately informed Wheat
of that fact. At most, there occurred a momentary conference between Officer
Samuelson and Officer Anderson about the latter’s previous encounters with the
driver. The district court did not err, let alone commit plain error, by denying
Wheat’s motion to suppress.




                                       -25-
B.    The Apprendi Error Was Harmless

       Wheat next claims that his sentence of 110 months is unconstitutional in
light of the Supreme Court’s decision in Apprendi v. New Jersey, 

(2000). In Apprendi, the Supreme Court held that “[o]ther than the fact of a prior
conviction, any fact that increases the penalty for a crime beyond the prescribed
statutory maximum must be submitted to a jury, and proved beyond a reasonable
doubt.” 

We have consistently extended this holding to federal drug
cases applying 21 U.S.C. § 841 (1994), the statute criminalizing possession with
intent to distribute controlled substances. See, e.g., United States v. Aguayo-
Delgado, 

, 932-33 (8th Cir. 2000), cert. denied, 

(2000); United States v. Sheppard, 

, 767 (8th Cir. 2000), cert. denied,

(2001).

       Wheat’s indictment charged him with possession with intent to distribute
more than 50 grams of cocaine base under §§ 841(a)(1) and (b)(1)(A), but the jury
acquitted him of that charge and convicted him of the lesser included offense of
simple possession of cocaine base under 21 U.S.C. § 844(a) (1994) instead.
Although this is the first time we have had to consider the application of Apprendi
to § 844, the task is functionally no different than with respect to § 841. A
conviction of simple possession of a controlled substance under § 844 may be
punished by a term of imprisonment not to exceed one year, or two years if the
individual, like Wheat, has previously been convicted of any drug offense under
any state law or under Chapter 13 of Title 21 of the United States Code. 21 U.S.C.
§ 844(a). However, § 844 also provides special enhanced penalties where the
controlled substance in question is cocaine base and certain specified quantities
are possessed, depending on the existence and number of prior convictions under
§ 844(a). See 

such as Wheat who have no prior convictions
under § 844(a), the statute provides that possession of cocaine base in an amount
exceeding five grams is punishable by a term of imprisonment not less than five

                                       -26-
years and not more than twenty years. See 

that because the jury
was not instructed to make a finding with respect to drug quantity, his sentence of
110 months considerably exceeds the two year maximum permitted under § 844(a)
where no quantity of cocaine base is determined.

      Although Apprendi was handed down after Wheat’s conviction and
sentencing, because his case is on direct appeal, we must apply it retroactively.
United States v. Anderson, 

, 429 (8th Cir. 2001) (citing Griffith v.
Kentucky, 

, 328 (1987)), cert. denied, 

(2001).
However, because Wheat “agreed that the quantity of [cocaine base] that he was
alleged to possess was [63.03 grams], and did not object to the admission of such
evidence at trial or to the court’s imposing a sentence based upon its own finding
of quantity, our review is for plain error.” United States v. Poulack, 

,
937 (8th Cir. 2001), cert. denied, 

(2001). Under that standard,

      before an appellate court can correct an error not raised at trial, there
      must be (1) error, (2) that is plain, and (3) that affects substantial
      rights. If all three conditions are met, an appellate court may then
      exercise its discretion to notice a forfeited error, but only if (4) the
      error seriously affects the fairness, integrity, or public reputation of
      judicial proceedings.

Johnson v. United States, 

, 466-67 (1997) (citations, brackets, and
internal quotation marks omitted); see also Fed. R. Crim. Pro. 52(b).

       We begin our analysis by assuming, as the government concedes, that plain
error has occurred.14 Wheat was properly charged with a specific quantity of drugs


      14
          We note, however, that in a case presenting nearly identical factual
circumstances, the Fourth Circuit held that no Apprendi error whatsoever had
occurred, reasoning that the jury had made a quantity finding. In United States v.
Brooks, 7 Fed. Appx. 248, 252 (4th Cir. 2001) (per curiam), cert. denied, 122 S. Ct.

                                         -27-
in the indictment. However, the jury was apparently not instructed to consider the
quantity charged as an element of the offense,15 and the relevant quantity was
therefore determined only at sentencing. Because failure to submit the element of
quantity to the jury would be error today, it was error then, and the first prong of
plain error review is satisfied. See 

. The second element
of the test is likewise met. “[I]n a case such as this--where the law at the time of
trial was settled and clearly contrary to the law at the time of appeal--it is enough
that an error be ‘plain’ at the time of appellate consideration.” 

261 (2001), the court considered the appeal of an individual arrested while in
possession of 13.11 grams of crack and, like Wheat, indicted under § 841(b)(1)(A),
but convicted only of the lessor included offense under § 844(a). The appellant was
charged in his indictment with possession with intent to distribute more than five
grams of crack. 

were instructed that to convict him of the lesser
included offense they had to find beyond a reasonable doubt that he had “possessed
the controlled substance described in the indictment,” the court held that the
instruction required the jury to find beyond a reasonable doubt that the appellant
possessed more than five grams. 

jury had effectively made a
specific finding on the element of quantity, and the conviction accordingly satisfied
Apprendi. 

instant case, Wheat’s indictment charged him with possession with intent
to distribute more than 50 grams, and the jury instruction with respect to the lesser
included offense was identical to the one in Brooks--i.e., the jury had to find that
Wheat had “intentionally possessed the controlled substance described in the
indictment.” Gov’t App., at 82, Final Instruction No. 18--Lesser Included Offense.
Brooks is therefore directly on point. However, expressing neither approval nor
disapproval, we decline to follow it at this time, as the law of our own circuit provides
sufficient grounds on which to affirm Wheat’s conviction.
      15
          The verdict form is simple, not special. The record on appeal does not
include a copy of Jury Instruction No. 17, Elements of the Offense. However, the
government has conceded that the jury did not make a particular finding with respect
to quantity.

                                          -28-
       The government argues that Wheat cannot pass the third and fourth prongs
of the clear error test. It relies on decisions of other circuits, e.g., United States v.
Swatzie, 

(11th Cir. 2000), cert. denied, 

(2001), as
well as our own decision in Anderson, 

, decided under the closely
related harmless error standard, wherein Apprendi errors have been held harmless
where overwhelming evidence of the quantity was adduced at trial, such that no
rational jury could have reached a verdict of conviction and yet not found that
quantity of drugs had it been charged to do so. Because the only drugs at issue in
this case were the 63.03 grams of cocaine base found in the McDonald’s bag, and
because Wheat did not contest that quantity of drugs at trial, see Trial Tr., at 64-
69, and repeatedly conceded its accuracy at the sentencing hearing,16 no rational
jury could have convicted him of possession of cocaine base except by reference
to the 63.03 grams of cocaine base. See 

.

       At oral argument, Wheat pursued a novel line of argument. Returning to a
theme he had adopted at the sentencing hearing, he argued that the Apprendi error
cannot be held harmless precisely because the jury that tried him was not rational.
Specifically, Wheat argues that 63.03 grams can only be considered to be a
distribution-level quantity of cocaine base, yet the jury expressly declined to
convict him of possession with intent. It therefore acted irrationally, and the
verdict arrived at could only be a compromise or a form of nullification. In these
circumstances, Wheat argues, the court simply cannot presume to state what the
jury would have done had it been instructed to consider quantity as an element of
the offense.

      We cannot accept this admittedly creative argument for two reasons. First,
we do not think that the jury verdict was irrational. In order to prove a violation of


      16
         See, e.g., Sentencing Tr., at 5 (responding affirmatively to court’s question
that quantity was 63 grams), 28 (“There’s no doubt that it was 63 grams.”).

                                          -29-
§ 841(a), the government had to prove that Wheat had the specific intent to
distribute the cocaine base. United States v. Franklin, 

, 998 (1984).
Although intent to distribute may be inferred solely from possession of a large
quantity of a controlled substance, id.; United States v. Gonzalez-Rodriguez, 

, 951 (8th Cir. 2001), the government always bears the burden of proof on
this issue. 

. In Franklin, we reversed a
conviction for possession with intent to distribute where the defendant did not
dispute that he knowingly and intentionally possessed 35 grams of 42 percent pure
cocaine at the time of his 

. We held that as a matter of law,
because the cocaine was not packaged in a manner consistent with distribution,
and the government offered no evidence of distribution paraphernalia, amounts of
cash, weapons, or other indicia of distribution, the evidence of mere possession of
35 grams was insufficient to support an inference of possession with intent to
distribute. 

We therefore remanded with instructions to enter a
judgment for simple possession under § 844(a). 

        We have scrutinized the trial transcript in the instant case and have found
the government’s evidence on the issue of intent to distribute to be
underwhelming. The only extrinsic evidence the government produced was that
Wheat was in possession of $250 in cash at the time of his arrest. See Trial Tr., at
29. On cross-examination of the government’s witnesses, Wheat’s counsel
emphasized the point that no drug notes or drug paraphernalia were seized. See

51. Nor, as Wheat’s counsel observed out of the hearing of the jury,
did the government’s witnesses testify that the amount of cocaine seized was a
quantity typically associated with distribution. See 

The evidence
presented showed only that Wheat possessed four large chunks of cocaine base,
not packaged in distribution-level quantities. Although the jury could have
inferred that 63.03 grams is a quantity sufficiently large that Wheat intended to
distribute it, we can hardly aver that it was irrational not to have done so. Indeed,
Wheat’s jury is not unique in acquitting under § 841(a) but convicting under §

                                        -30-
844(a) where the quantity would have supported a determination either way. See,
e.g., Brooks, 7 Fed. Appx. at 249-52 (13.11 grams of crack).

        Moreover, even if we suspected that the jury had reached a compromise
verdict, the Supreme Court does not direct us to try to go into the mind of the jury
to glean evidence of irrational jury behavior. In Neder, the Court framed the
question for the reviewing court thus: “Is it clear beyond a reasonable doubt that a
rational jury would have found the defendant guilty absent the 

(emphasis added). Indeed, guessing about the jury’s rationale is precisely what
courts must not do. We must accept the verdict that the jury handed down, which
was a conviction on the count of simple possession of cocaine base, and the only
evidence of cocaine base in this case was the 63.03 grams in the McDonald’s bag.
Wheat never contested the accuracy of the quantity measurement at trial, and
repeatedly conceded its accuracy at sentencing. As no rational jury could have
convicted Wheat of possession of cocaine base yet found him responsible for less
than five grams, any Apprendi error did not affect Wheat’s substantial rights; nor
did it seriously affect the fairness, integrity, or public reputation of his trial. See

(holding that defendant’s substantial rights were
unaffected where quantity was charged in the indictment, the defendant stipulated
to quantity at trial, and defendant’s counsel conceded accuracy of court’s
determination of quantity at sentencing); United States v. Nelson, 9 Fed. Appx.
583, 584 (8th Cir. 2001); but cf. United States v. Frazier, -- F.3d --, 

, at *14 (8th Cir. Nov. 6, 2001) (holding that although defendants’
substantial rights were affected, “[b]ecause the evidence overwhelmingly
establishes a quantity of drug sufficient to authorize the sentences imposed, we
conclude the jury’s failure to find the amount of heroin did not seriously affect the
fairness, integrity, or public reputation of the proceedings”); United States v.
Soltero-Corona, 

, 860 (8th Cir. 2001).




                                         -31-
III.   CONCLUSION

      The district court properly denied Wheat’s motion to suppress, and any
Apprendi error is harmless. We have considered Wheat’s remaining arguments
and find them to be without merit. The judgment of conviction is affirmed.

       A true copy.

             Attest:

                      CLERK, U.S. COURT OF APPEALS, EIGHTH CIRCUIT.




                                     -32-