In this appeal the state challenges the holding of the Court of Appeals, upon which it ordered a new trial, that the trial court erred in admitting into evidence a boat charter agreement having a signature, “Milan LeDuc” affixed as “charterer,” and then permitting the jury to compare known samples of defendant’s handwriting with the signature on the charter agreement without the aid of competent opinion testimony.
The Court of Appeals interpreted G.S. 8-40, which deals with “proof of handwriting by comparison,” and our rules of evidence, to prohibit handwriting comparisons by a jury unaided by competent lay or expert testimony. Defendаnt argues that the Court of Appeals was correct in holding that the charter agreement was *64 improperly admitted. In addition, defendant makes several cross-assignments of error under Rule 10(d) of the Rules of Appellate Procedure. The most significant assignment of error is that the evidence is insufficent to support the verdict; 2 therefore, argues defendant, the Court of Appeals erred in sustaining the trial court’s denial of his motion to dismiss.
We decide that the Court of Appeals incorrectly concluded that the jury could not compare the handwriting samples without the benefit of opinion testimony, and that it was also error to sustain thе trial court’s denial of defendant’s motion to dismiss. Defendant’s conviction is, therefore, reversed.
The state’s evidence tends to show as follows:
On 26 April 1977, a man who identified himself as Milan A. LeDuc (herein “LeDuc”) chartered the Frances Ann, a sixty-five-foot fishing trawler, from its owners, two commercial fishermen, Andrew Tiner and Charles Daniels, in Fort Myers Beach, Florida. The trawler was then located at Bayou La Batre, Alabama. Tiner and Daniels had purchased the trawler on 14 April 1977. “LeDuc” had on 14 April first approached Tiner and Daniels at Bayou La Batre and told them he had earlier wanted to buy the trawler and wished to charter it. At a later meeting in Tampa, Florida, the parties agreed оrally on a charter agreement which was consummated in writing on 26 April in Fort Myers Beach. The charter provided for a term of three months at a rental of $3500 per month, the first month’s rent payable in advance. It also called for a $1000 security deposit. “LeDuc” paid only the $1000 security deposit in cash. Neither Tiner nor Daniels saw “LeDuc” sign the charter. “LeDuc” brought the charter to them at some point in the negotiations with the signature “Milan LeDuc” already affixed. Upon the request of Tiner and Daniels for some identification, “LeDuc” produced a Florida driver’s license bearing the name “Milan A. LeDuc” and a photograph that aрpeared to be *65 that of “LeDuc.” At all these meetings “LeDuc” was accompanied by the same two companions who were introduced only by their first names, which Tiner and Daniels could not recall. “LeDuc” told Tiner and Daniels that he wanted the trawler for sponge diving. Daniels had noticed diving equipment in a truck driven by “LeDuc.” Neither Tiner nor Daniels could identify defendant as being “LeDuc.” Daniels testified that defendant was “not the same man”; there was a “likeness, that’s all.” According to Tiner and Daniels, the hold of the boat was freshly painted when they bought it and there was nothing to indicate that marijuana had ever been aboard.
On 18 May 1977, betweеn 2 and 2:30 a.m., the Frances Ann was observed moored at the dock at Stumpy Point, Dare County, North Carolina, by Raynor Twiford, a commercial fisherman who lived nearby. The boat had not been there at 10 p.m. on 17 May. Twiford observed an unidentifiable truck approach “from towards Engelhard on U.S. 264.” He heard noises “similar to what it would be if they were unloading something off a boat.” The truck remained in position “20 minutes at the most,” then “pulled out . . . headed North on Highway 264.” After this truck left the scene Twiford observed three unidentified persons come “from the direction of the Frances Ann,” leave the dock and go to a nearby unlocked building where they remained about five minutes. These persons then returned to the dock and left in what sounded like a second pickup truck. Twiford could not say whether these persons were men or women.
On 22 May 1977, Leland Wise, a long-time Stumpy Point resident, who had also been observing the Frances Ann since 18 May, reported its presence at the dock to the sheriff. Wise had observed that the Frances Ann, although a fishing trawler, was not rigged for fishing and was not moored in a manner that would adequately secure the boat to the dock. There were no spring lines; there were only one light stern line and one light bow line used to moor the trawler. The door and windows to the deck-housе were open and the lights were left on.
Investigation of the vessel by law enforcement officers revealed the following: It was unoccupied. Twenty-two and four-tenths pounds of marijuana were discovered in the captain’s quarters. Of some thirty latent fingerprints lifted from the vessel *66 and its contents, eighteen were defendant’s, one was that of an investigating deputy sheriff, and the others could not be identified. Defendant’s prints were found on a Caribe 55 radio, a coffee cup located in the wheel room just to the right of the wheel, a coffee can located on a cupboard by the stove, a coffee pot loсated on the stove in the galley, an Aunt Jemima pancake box located on a cupboard in the galley, an oil can, a cardboard Duracell battery box, a Hormel sausage wrapper located in a trash can, and a blank page in a navigational notebook. A small quantity of meat inside the sausage wrapper appeared “fairly fresh.” Because of the high humidity which gives fingerprints a “shorter life span” and the nature of “the articles they were on,” the fingerprint expert’s opinion was, “the prints could not have been there for a very long period of time.” The maximum time was “two weeks.” 3 The vessel was equipped with navigational equipment suitable for the high seas. Marijuana seeds and green vegetable debris were found in the hold areas. Burlap bag indentations were located on the bulkheads rising five to six feet in two larger holds, and three to three and one-half feet in a smaller hold. The combined volume of the holds was 370 cubic feet. Three bunks on board appeared to have been slept in, and there were three soiled plates in the galley.
From entries in a navigational notebook and various charts found on board, it appeared that the Frances Ann was positioned in Mobile Bay, Alabama on 28 April 1977, then traveled to an area off the north coast of Colombia and Venezuela, and then up the east coast of the United States to Stumpy Point.
Finally, the state offered the charter agreement with the signature “Milan LeDuc” affixed thereto as “charterer,” three *67 samples of defendant’s handwriting and signature obtained from him by investigating officers, and a court document bearing defendant’s signature, all for the purpose of the jury’s “comparing the signatures on the five separate documents pursuant to North Carolina General Statutes 8-40.” This Court has examined these documents. In our opinion the known signatures of defendant are sufficiently similar to that on the charter agreement to permit a jury reasonably to infer that defendant signed the charter agreement.
Defendant offered no evidence, but moved to dismiss for insufficiency of the state’s evidence. This motion was denied.
The jury acquitted defendant of the charge of possession of marijuana but convicted him of the charge of conspiracy to possess marijuana.
I.
The first question we address is whether at trial the fact-finder may compare a known sample of a person’s handwriting with handwriting on a contested document and thereby determine whether the handwriting is the same on both without the aid of competent lay or expert testimony. The Court of Appeals, in
State v. Simmons,
In order to understand current North Carolina law on handwriting comparison by the jury, it is necessary to emphasize that historically three separate questions have arisen. First, there is the question of the competenсy of a given witness to express an opinion on the genuineness of a contested writing. Second, there is the question of what may serve as a standard of comparison. Finally, there is the question of the right of the jury to compare a standard with a contested writing. Unfortunately, the Court has inadvertently confused these three questions in several cases.
Martin v. Knight,
*68
Originally, a person was permitted to give his opinion on the authenticity of a contested writing if he had “seen the party write or [had] obtained a knowledge of the character of his writing from a correspondence with him upon matters of business or from
transactions
between them.”
Pope v. Askew,
The group of lay people allowed to testify also has been held to include those persons whose familiarity with an individual’s handwriting was based on observation of the author’s handwriting in ancient documents. Such a witness is allowed to testify on the authenticity of a document even though he is not an expert in handwriting analysis if he has had “full opportunity and frequent occasion to examine them.”
Nicholson v. Lumber Co.,
Subsequently, a third type of witness, the expert, was allowed to testify on the authenticity of a given handwritten document if he qualified because of his skill in handwriting analysis.
Yates v. Yates,
*69 The expert witness, unlike the lay witness, need not have any familiarity with the handwriting at issue before the beginning of the controversy. At trial, he compares the handwriting on the contested document with a genuine standard. Based on this comparison he gives his opinion on the authenticity of the contested document.
The question of what may serve as a standard for comparison was originally answered very narrowly. Only writings received in evidence and admittedly genuine or which the party whose handwriting is at issue was estopped to deny or had conceded were genuine could be used as exemplars.
Tunstall v. Cobb, supra,
The question of what the jury may see when the authenticity of a document was challenged was originally answered so as to prohibit jury examination of the documents either during or after the witness testimony.
See, e.g., Fuller v. Fox,
It was supposed in the past that the average juror was not sufficiently intelligent — educated—to comprehend the fine *70 shades of difference in handwriting. Whatever may be thought of the soundness of the reason in the past, it is manifest that it has but little force at this time. As education and intelligence have increased and the methods of illustration improved, the capacity of the ‘average man’ to write and pass upon the handwriting of others has advanced.
Thus, the Court concluded that an opinion witness, expert or not, should be allowed to show the disputed and genuine writings to the jury and explain why he thinks there was or was not a difference in the writings.
Id.
at 577-78,
In 1913 our legislature joined those оf a number of states with statutes governing handwriting comparisons when it adopted language substantially similar to that used in the English statute. Compare Act of Mar. 5, 1913, ch. 52, 1913 N.C. Sess. Laws 98 with The Common Law Procedure Act, 1854, 17 & 18 Vit., ch. 125, § XXVII. Our statute, currently codified as G.S. 8-40, states:
In all trials in this State, when it may otherwise be competent and relevant to compare handwritings, a comparison of a disputed writing with any writing proved to the satisfaction of the judge to be genuine, shall be permitted to be made by witnesses, and such writings and the evidence of witnesses respecting the same may be submitted to the court and jury as evidence of the genuineness or otherwise of the writing in dispute ....
This statute has been interpreted several times by this Court since its passage аs changing the common law in several respects.
*71
In
Bank v. McArthur,
In
Newton v. Newton,
Subsequent cases have recognizеd that the statute liberalized the common law in some respects and have tended to answer specific questions permissively rather than restrictively. For example, in
Gooding v. Pope,
In re Will of McGowan,
Thus we reach the question raised in the instant appeal. In our view the plain words of G.S. 8-40 do not prohibit handwriting comparisons by the jury without the aid of opinion testimony. The statute states “a comparison of a disputеd writing with any writing proved to the satisfaction of the judge to be genuine, shall be permitted to be made by witnesses, and such writings and the evidence of witnesses respecting the same may be submitted to the court and jury.” (Emphasis supplied.) The statute does not mandate that writings be submitted only with evidence of witnesses; it merely states that writings and testimony may be submitted to the trier of fact as evidence of the authenticity of a contested document.
Even though the statute does not expressly prohibit comparison, it is clear that under our common law rules of evidence the jury was not allowed to compare handwritings without the aid of testimony.
Martin v. Knight, supra,
We believe there are sound reasons for modifying the common law rule at issue. During the period in which this rule developed, many jurors had little or no education and little daily experience with handwriting comparison. But it was noted in
Martin v. Knight, supra,
Whatever may have been the experience and competence of common-law jurors to assess the genuineness of signatures, we are of the opinion that this aptitude is one which today most laymen have been obliged to develop in conducting their own affairs. With the widespread use of credit cards and travelers’ checks, merchants and others in the field of commerce are frequently confronted with the necessity of comparing signatures. In the light of this common experience and exposure, we hold that a factfinder may, in the discretion of the court, be permitted to resolve the issue of forgery without expert assistance. Under our law it is not incumbent on jurors to accept an expert’s opinion blindly. They must come to their conclusion on the basis of their own observations and experience and assessment of all the evidence before them.
In short, the average juror today, through increased education and experience, is as prepared to compare handwriting as many witnesses qualified as experts in earlier cases.
See, e.g., Yates v. Yates, supra,
Furthermore, under present rules, common law and statutory, the jury is permitted to compare documents in evaluating and weighing the witness’ testimony, bоth expert and lay. The jury’s determination of the weight it gives such testimony, particularly when testimony conflicts, must be colored by the jury’s own comparison of the various documents before it.
Finally, a number of jurisdictions, either by statute or case law, have made the decision to allow a jury to compare writings unaided by testimony.
See, e.g., Parker v. State,
Before handwritings can be submitted to the jury for its comparison, however, the trial judge must satisfy himself that one of the handwritings is genuine. The statute so provides. We hold, in *74 addition, that the trial judge must also be satisfied that there is enough similarity between the genuine handwriting and the disputed handwriting, that the jury could reasonably infer that thе disputed handwriting is also genuine. Both of these preliminary determinations by the trial judge are questions of law fully reviewable on appeal. In the instant case, the samples shown to the jury for comparison with the disputed charter were given by the defendant himself. Having examined these samples with the disputed signature on the charter, we are satisfied that there is enough similarity between them for the documents to have been submitted to the jury for its comparison.
Thus, we conclude that the charter agreement was properly submitted to the jury for comparison with handwriting exemplars executed by defendant.
II.
We now turn to the question whether there was sufficient evidence that defendant conspired with others to possess 22.4 pounds of marijuana to survive defendant’s motion to dismiss.
The test of the sufficiency of evidence in a criminal case has been articulated by the United States Supreme Court as whether, “after viewing the evidence in the light most favorable to the prosecution,
any
rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.”
Jackson v. Virginia,
The evidence is to be considered in the light most favorable to the State; the State is entitled to every reasonable in-tendment and every reasonable inference to be drawn therefrom; contradictions and discrepancies are for the jury *75 to resolve and do not warrant dismissal; and all of the evidence actually admitted, whether competent or incompetent, which is favorable to the State is to be considered by the court in ruling on the motion.
In addition to producing substantial evidence of each of the material elements of the particular offense, the state must produce substantial еvidence that the defendant committed it.
In re Vinson,
A criminal conspiracy has been defined in this state as “an agreement between two or more persons to do an unlawful act or to do a lawful act in an unlawful way or by unlawful means.”
State v. Abernathy,
Although no overt act or express agreement is required, the state must prove defendant’s “intent to accomplish some crime or unlawful purpose, or to bring about some end, not in itself criminal or unlawful, by criminal or unlawful means.”
State v. Wrenn,
In order to be guilty as a conspirator, therefore, it is not necessary that a defendant “personally participate in the overt act,” but it must “be established by competent evidence that he entered into an unlawful confederation for the criminal purposes alleged.”
State v. Andrews,
Permissible inferences, moreover, do not include those based upon other inferences. “[C]harges of conspiracy are not to be made out by piling inference upon inference, thus fashioning what [has been] called a dragnet to draw in all substantive crimes.”
Direct Sales Co. v. United States,
Although it is not necessary for the state to prove defendant personally possessed marijuana
6
in order to convict him of conspiracy to possess marijuana,
see, e.g., State v. Andrews, supra,
The Court of Appeals concluded,
Evidence that after a long sea voyage thе boat docked at night at an isolated point in Dare County and was almost immediately thereafter met by persons who arrived at the scene in two trucks, furnishes solid support for the inference that the meeting took place by prior agreement. Indeed, it seems almost inconceivable that such a meeting could have occurred without prior arrangement. The inference that the purpose of the meeting was to unload marijuana from the boat into one or both of the trucks is equally solidly supported by the evidence.
We agree that there is direct evidence that a boat with marijuana aboard was met shortly after its arrival at a point in Dare County by unknown persons and that from the fact of such a meeting, a jury could infer that someone had pre-arranged it. The difficulty in the state’s case is with the question, “Who arranged the meeting and with whom?” The Court of Appeals said, “that defendant was one of the persons who joined in making the agreement may be reasonably inferred from the evidence that he had chartered and had participated in navigating the boat during the voyage in question.”
Id.
at 236,
The Court of Appeals seems to have overlooked the proposition that there is no direct evidence that defendant chartered the boat, participated in its navigation, or was ever aboard at the time marijuana was being transported. The jury could, of course, reasonably infer from other evidence adduced that all of these things were true. It could infer, from similarity in the signatures and names and the Florida driver’s license shown to the owners of the trawler, that defendant was the same Milan LeDuc who arranged for and executed the charter. It could infer from defendant’s fingerprints found on board the vessel, the places where these prints were found, and defendant’s Coast Guard license application that defendant had participated in navigating the trawler and was on board at the time marijuana was being transported. It is only by building on these inferences, however, *78 that the jury might then further infer that defendant participated in an unlawful agreement to possess marijuana.
The state is thus met head-on by our rule that in circumstantial evidence cases inferences may not be built upon inferences in order for the fact-finder to reach the ultimate facts upon which guilt must be premised.
Direct Sales Co. v. United States, supra,
In Parker, defendant was convicted by a jury of breaking and entering and larceny of clothing from Robert Hall Clothing Store in Charlotte. The state relied on the doctrine of recent possession. It offered direct evidеnce that Robert Hall’s front glass doors had been broken through on the evening of 28 January 1966. There was blood on the floor inside the doors. From an inventory taken four days before this breaking, it was determined that five suits were missing from the store. At approximately 11 p.m. on the evening of 28 January, a witness observed a man approximately one block from the store drop something and run. The witness said this man “looked just like the defendant” but “I am not for sure that this defendant was the man” and “[I] cannot say beyond a reasonable doubt that the defendant was the man.” The item dropped, however, was one of the stolen suits. Shortly after the witness picked up the suit, defendant walked up to him from the direction in which the unidentified person had run. Defendant’s hand was cut.
This Court held in
Parker
that the evidence was insufficient to support the verdict on the theory of recent possession because there was no “direct evidence” that defendant ever recently possessed the stolen property. The Court said,
There was no direct and clear evidence placing the stolen goods in the possession of defendant.
‘A basic requirement of circumstantial evidence is reasonable inference from established facts. Inference may not be based on inference. Every inference must stand upon some clear and direct evidence, and not upon some other inference or presumption.’ [Citations omitted.]
*79 In other words, the Court in Parker concluded that since the jury must first infer defendant’s recent possession of the stolen property and from that inference further infer that he was the thief, the evidence was insufficient to carry the case to the jury. So it is here.
The decisions of the Court of Appeals concluding (1) that it was error to admit the signature on the charter agreement and other samples of defendant’s handwriting for the jury’s comparison, and (2) that the evidence was sufficient to be submitted to the jury on the conspiracy charge аre, therefore,
Reversed.
Notes
. Defendant also contends that the evidence used at trial was the product of an illegal search and should have been excluded; the admission of an application to the Coast Guard for a boat operator’s license purportedly made by defendant violated certain of defendant’s constitutional rights; the district attorney argued improperly to the jury about defendant’s failure to testify and corrective instructions were inadequate; testimony of a state’s witness was incompetent; and, finally, the jury instructiоns did not sufficiently define the element of willfulness in a conspiracy charge and failed to give equal stress to the contentions of each side.
. On cross-examination the witness testified as follows:
Q. Approximately two weeks? It could be three weeks?
A. Yes, sir.
Q. It could be one week?
A. Yes, sir.
Q. It could be four weeks?
A. Yes, sir, it could.
Q. It could be one day?
A. Yes, sir.
.
See also Nicholson v. Lumber Co., supra,
. For a clear and accurate summary of different articulations of what essentially is the same test and a conclusion that the better articulation is that there must be “substantial evidence” of each element of the crime, see
State v. Smith,
. We need not discuss the question whether the evidence was sufficient to convict defendant of possession of marijuana. The jury acquitted defendant of that offense.