Well-Made Toy Mfg. Corp. v. Goffa International Corp.

MEMORANDUM, ORDER AND JUDGEMENT WEINSTEIN, Senior District Judge. TABLE OF CONTENTS: I. Introduction.153 II. Procedural History.153 III. Facts. 153 A. Rag Dolls. 153 B. ' Sweetie-Mine .153 C. Huggable-Lovable.155 IV Law. .157 A. Jurisdiction. v-H 1. Registration Requirement .... tH 2. As Applied to Derivative Works i — ( B. Infringement. i — I 1. Valid Copyright. t — l 2. Copying of Original Elements . i — 1 a. Actual Copying. i — | b. Illegal Copying. r — | i. Protectible Elements. i — I ii. Substantial Similarity i — ( V. Application of Law to Facts. CD i-l A. Registration. ^ ÍD i — t B. Infringement. 10 CD i — I 1. Valid Copyright. 10 tH *153 2. Copying of Original Elements. lO CO 7 — 1 a. Actual Copying. io CD 7 — 1 b. Illegal Copying. CD 7 — 1 VI. Conclusion. .172 I. Introduction Based on its copyright and registration of its 20 inch ragdoll, plaintiff sues for copying of its 48 inch rag doll, designed by it as a grown-up version of the smaller toy. Judgement for defendant is required because small dolls, like humans, grow up with substantially changed proportions. II. Procedural History In 1999, plaintiff, Well-Made Toy Manufacturing Corporation (“Well-Made”), filed separate actions against Goffa International Corporation (“Goffa”), the manufacturer of the allegedly infringing doll, and King Kullen Grocery Company (“King Kullen”), a chain of supermarkets that sold it at retail. Since Well-Made has indemnified King Kullen, the singular word “defendant” is used throughout this memorandum. The cases were consolidated for all purposes. They were tried without a jury. III. Facts A. Rag Dolls Rag dolls are soft, stuffed fabric playthings which were traditionally made at home from scrap pieces of fabric. Currently, most rag dolls are manufactured by toy companies. The doll designer in this country creates annotated paper patterns. Mass cutting of individual fabric pieces and sewing together of the parts to create an empty shell, or “skin,” are typically accomplished in China. The shapes of the constituent fabric pieces and the way in which the pieces are assembled and stuffed determines the three-dimensional form of the doll. The skins are stuffed in this country to give the toy volume. Designing rag dolls is not a field in which experimentation or deviation from the norm is highly valued. The standard elements are well-known. As plaintiffs designer testified, “it is putting things together to make one new creation that makes it individual.” Rag dolls typically have in one skin a head, a torso, limbs, hands, and feet; a dress and bonnet can either be attached or separate. They commonly incorporate quaint outfits, including dresses of traditional style, print fabric with matching bonnets, hair consisting of yarn which is often braided, faces which often have embroidered features, limbs that are sometimes more floppy than the torso, and shoes which are built in and stuffed with the same material as the rest of the doll. Ruffles, cuffs, and bows are common features. Density of the stuffing determines rigidity. B. Sweetie-Mine The plaintiff, Well-Made, produces a ragdoll named “Sweetie Mine”. It was designed for plaintiff in 1995 by Marla Speer. Ms. Speer began by drawing a two-dimensional illustration. She then created fabric patterns for the pieces that would make up the three-dimensional doll. She also laid out embroidery for the face and painted a pattern for the doll’s dress. To give a more defined form to the doll’s face, the designer incorporated a polyester backing, or “face pocket.” She testified that she borrowed the idea for a “face pocket” from generally available designs of teddy bears and other stuffed animals. Her doll was 20 inches tall. *154 Well-Made received a copyright registration for the Sweetie Mine doll in 1996. Copyright Registration No. VA 770-506 (dated April 26, 1996). The deposit pictures accompanying the copyright application were photographs of the 20 inch doll. Copyrightable features of the work were identified as the “facial artwork and body sculpture.” In early 1998, Well-Made developed a 48 inch version of the Sweetie Mine. This 140 percent increase in scale was accomplished by using a photocopy machine to enlarge the paper drawings of the 20 inch fabric patterns. Enlarged patterns were then adjusted so that the larger doll’s proportions would remain pleasing. For example, the head was reduced in size and the legs lengthened. Ms. Speer testified that these adjustments were intended to preserve the “same aesthetic appeal as the original design.” Krista Kosmas, who also participated in the creation of the 48 inch doll for Well-Made, testified that enlarging the doll without altering the proportions would result in a “fat, very klutzy looking doll.” Both admitted that these adjustments were not mathematically governed by any strict rules of proportionality; artistic discretion was required. Ms. Speer’s testimony was as follows: Q: Was there an exact ratio of the head to the body of the forty-eight inch doll that you were working toward? A: No. Q: There was just whatever you thought looked good? A: Yes. Q: In your artistic judgement? A: Yes. Q: Were there a range of sizes for the head in comparison with the body of the forty-eight inch doll that you might have chosen? A: I think there might be a small range. Q: Some other designers might have chosen a different proportion of the head to the body and the rest of the doll? A: Oh, yes. Might not have looked as good, though. Aside from these proportional considerations, three substantive changes were made that seem unrelated to proportion. First, the shape of the faces is different. That of the 20 inch doll has five distinct angles: two just below the cheeks, two at the temples, and one at the peak of the forehead. This gives the face a soft pentagonal look, like an inverted strawberry that comes to a peak at the forehead. The 48 inch doll has a wider, more elliptical face; the distinct angles are absent; and the head juts forward from the body further as a result of larger darts in the fabric. Second, the features of the 20 inch doll are set lower in the face, so that it has a high forehead; the facial features of the 48 inch doll are more centered, so that the forehead is lower. These changes give the 48 inch doll an “older” look than the 20 inch doll. Finally, the bow, or sash, which was included on the back of the waistband of the 20 inch doll was omitted from the 48 inch product. All in all, the Well-Made 48 inch rag doll is an attractive, perky looking four foot companion for a preteen child, while the 20 inch is sweet and cuddlesome, small enough to be snuggled up to by a toddler. Well-Made did not file a copyright application for registration of the 48 inch doll. *155 C. Huggable-Lovable The defendant manufactures and imports toys. It has been in the business of selling ragdolls since 1992. Among the dolls sold by Goffa was a small seated ragdoll (CL), which it copyrighted in 1996. Douglas Song, President of defendant Goffa, learned that a four foot tall rag doll (plaintiffs) was selling well in United States markets. He decided to introduce a competitive work. In April of 1998, he ordered his Shanghai factory to develop a prototype based on the face and hat of his CL Doll, and body parts from various other Goffa dolls. This factory had never manufactured rag dolls. Edward Sclier, defendant’s general manager, testified that the factory employees were “not creative” when it came to product design, and that they “couldn’t do much on their own.” Later that month, Mr. Song and Mr. Sclier attended a trade fair in China. There they noticed that a company named Hubei was exhibiting rag dolls of differing sizes; one doll in Hubei’s catalogue appears to be a copy of plaintiffs 20 inch doll. Hubei was not displaying any 48 inch dolls at the fair, so Mr. Song requested that Hubei create a 48 inch sample and quote a price. Mr. Song testified that while at the fair he did not view the Hubei catalogue, but that seems doubtful since he was there to see what was available in the market. Defendant admitted obtaining a sample 48 inch doll from Hubei. The evidence suggested that the Hubei 48 inch sample doll’s face and body strongly resembled the Well-Made’s 48 inch Sweetie Mine. Mr. Song sent the 48 inch sample he had received from Hubei to the Shanghai factory as an example of “how nice other people make it.” Defendant also obtained at least one of Well-Made’s 48 inch dolls and disassembled it during its own design process. Mr. Sclier admitted that he himself had purchased a Well-Made 48 inch doll at retail in this country. While he swore otherwise, it is apparent from his equivocal statements on the stand that he sent this doll to the company’s Shanghai factory in order to have it reproduced; there, it was copied with some variations based on the Hubei 48 inch sample. Defendant’s Chinese factory made design changes based on the Hubei sample, some of Goffa’s own dolls, and the Well-Made doll. Mr. Song decided to use his own factory’s design, and informed Hubei that he would not be placing an order with it. As already suggested, and is apparent from the picture set out below, defendant’s 48 inch doll has an entirely different face than plaintiffs Sweetie-Mine 48 incher, with a head that does not stand out from the body as much as does plaintiffs; has slightly different body parts; uses dissimilar clothing, with a style copied only in part from plaintiffs doll’s dress (as in the ankle and the wrist bands); embodies a much longer skirt; and utilizes poorer cloth and sewing. The stuffing seems about the same in density. All in all, the alleged infringing 48 inch doll looks like a slightly less alert and attractive relative of plaintiffs 48 inch doll, but definitely not like a twin of either the 20 or 48 incher. *156 [[Image here]] Plaintiffs Sweetie Mine doll was sold in the United States by retailers including King Kullen Supermarkets, a defendant in this action. Defendant’s Huggable Lovable doll was introduced at a lower price than what plaintiff had been asking. To keep competitive, plaintiff then changed its 48 inch doll slightly by “slimm[ing] the pattern down,” cutting its cost, and lowering its price. In October 1998, attorneys for plaintiff warned defendant that they believed Gof-fa’s 48 inch doll infringed on copyrights owned by Well-Made. A few months later, in early 1999, Goffa instructed its factory to substantially change the design of its 48 inch doll and it ceased production of the allegedly infringing version. *157 IV. Law A. Jurisdiction 1. Registration Requirement For any work created after January 1, 1978, copyright automatically inheres upon the work’s creation. See 17 U.S.C. § 102(a) (“Copyright protection subsists, in accordance with this title, in original works of authorship fixed in any tangible medium of expression ... ”); Montgomery v. Noga, 168 F.3d 1282, 1288 (11th Cir.1999); 2 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 7.16[A][1] at 7-147 (2002). Registration with the United States Copyright Office is not required to obtain copyright protection. 17 U.S.C. § 408(a) (“At any time during the subsistence of [the copyright], the owner of copyright or of any exclusive right in the work may obtain registration of the copyright.... Such a registration is not a condition of copyright protection.”). Registration is a precondition for bringing an infringement action in federal court. 17 U.S.C. § 411(a) (“no action for infringement of the copyright in any United States work shall be instituted until registration of the copyright claim has been made in accordance with this title.”). It is a jurisdictional prerequisite, not a substantive element. See Geritrex Corp. v. Dermarite Indus., LLC, 910 F.Supp. 955, 966 (S.D.N.Y.1996). Actual certification by the Copyright Office is not necessary. A copyright holder may commence an action for infringement as soon as the Copyright Office has received a proper application, fee and deposit. See Apple Barrel Prods., Inc. v. Beard, 730 F.2d 384, 386-7 (5th Cir.1984); 2 Nimmer § 7.16[B][1][a] at 7-155. But see Tooker v. Copley, 3 U.S.P.Q.2d (BNA) 1396, 1987 WL 124315 (S.D.N.Y.1987) (plaintiff must obtain a response from Copyright Office; application alone is insufficient). 2. As Applied to Derivative Works A work not itself registered, but derived from a copyrighted registered work may be the basis of a suit. Section 10 of Title 17 of the United States Code defines a derivative work: A ‘derivative work’ is a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgement, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a ‘derivative work’. Derivative works receive copyright protection. 17 U.S.C. § 103. “A derivative work ... must incorporate that which itself is the subject of copyright.” 1 Nimmer § 3.01 at 3-5. A threshold of more than trivial originality is necessary for a work to be independently copyrightable. As noted in Durham Industries, Inc. v. Tomy Corp., 630 F.2d 905 (2d Cir.1980): First, to support a copyright the original aspects of a derivative work must be more than trivial. Second, the scope of protection afforded a derivative work must reflect the degree to which it relies on preexisting material and must not in any way affect the scope of any copyright protection in that preexisting material. Thus the only aspects of [a derivative work] entitled to copyright protection are the non-trivial, original features, if any, contributed by the author or creator of those derivative works. *158 Id. at 909. In Durham, the court of appeals for the Second Circuit found that toy-replicas of popular Disney characters failed to satisfy this test. The Durham court reiterated the rule established in L. Batlin & Son, Inc. v. Snyder, 536 F.2d 486 (2d Cir.1976) (en banc), cert. denied, 429 U.S. 857, 97 S.Ct. 156, 50 L.Ed.2d 135 (1976), that a work involving changes (such as reproduction in another medium) requiring only “manufacturing” or “physical” skill, as opposed to “artistic” skill, does not merit protection as a derivative work. Durham, 630 F.2d at 910; see also 1 Nimmer § 3.03[A] at 3-11-3-12. (“[T]he following contributions to pre-existing works have been held to be too minimal to warrant the recognition of a new and separate copyright as a derivative[:] ... a change in ... scale or size of a work or sculpture .... ”). Because derivative works receive copyright protection separate from that of the preexisting works they modify, a jurisdictional problem arises if the holder of a registered copyright creates a derivative work based on the registered work, fails to register the derivative work, and then brings an infringement claim based on the derivative work. The requirement in section 411 of Title 17 of the United States Code that copyright claims be registered before a lawsuit can be entertained makes no distinction between derivative and non-derivative works. The copyright in the preexisting work upon which a derivative work is based is not sufficient to meet the registration requirement as to any claim based on infringement of the derivative work. See Murray Hill Publ’ns, Inc. v. ABC Communications, Inc., 264 F.3d 622 (6th Cir.2001); Creations Unlimited, Inc. v. McCain, 112 F.3d 814, 816 (5th Cir.1997). This correct approach must be applied carefully in accordance with the admonition that “the scope of protection afforded a derivative work must reflect the degree to which it relies on preexisting material and must not in any way affect the scope of any copyright protection in that preexisting material.” Durham, 630 F.2d at 909. By definition, a derivative work contains elements of underlying works that themselves may be subject to copyright; the copyright in the derivative work extends only to those elements original to the derivative work. It follows that where the preexisting work is registered, but the derivative work is not, a suit for infringement may be maintained as to any protected element contained in the registered preexisting work, but not as to any element original to the unregistered derivative work. See Mattel, Inc. v. Robarb’s, Inc., 139 F.Supp.2d 487, 496, 498 (S.D.N.Y.2001) (defendant’s copy of plaintiffs unregistered derivative work infringed on plaintiffs copyright in registered work, elements of which were included in the unregistered derivative work and the infringing copy); 2 Nimmer § 7.16[B][2] at 7-165. B. Infringement To establish copyright infringement, “two elements must be proven: (1) ownership of a valid copyright, and (2) copying of constituent elements of the work that are original.” Feist Publ’ns. Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 361, 111 S.Ct. 1282, 1296, 113 L.Ed.2d 358 (1991). 1. Valid Copyright Assuming that a plaintiff satisfies the jurisdictional requirement by registering the copyright, the first substantive inquiry the court must make is whether the copyright is valid. The Copyright Act provides that a “certificate of [copyright] registration made before or within five years *159 after first publication of the work shall constitute prima facie evidence of the validity of the copyright....” 17 U.S.C. § 410(c). This presumption of validity may be rebutted. See Hasbro Bradley, Inc. v. Sparkle Toys, Inc., 780 F.2d 189, 192 (2d Cir.1985); Durham Indus., Inc. v. Tomy Corp., 630 F.2d 905, 908 (2d Cir.1980). 2. Copying of Original Elements The second element of the Feist test, that a plaintiff prove “copying of constituent elements of the work that are original,” requires a two part showing: “A plaintiff with a valid copyright proves infringement by demonstrating that: (1) the defendant has actually copied the plaintiffs work; and (2) the copying is illegal because a substantial similarity exists between the defendant’s work and the protectible elements of plaintiffs.” Fisher-Price, Inc. v. Well-Made Toy Mfg., 25 F.3d 119, 122-23 (2d Cir.1994) (emphasis in original). a. Actual Copying Copyright, unlike patent protection, is not solely concerned with the degree of similarity between the protected work and the allegedly infringing work. In copyright, the explanation for the similarity matters. For there to be a violation of copyright protection, similarities must arise from actual copying, not mere coincidence. Proof of independent creation is an affirmative defense against a claim of infringement. Yur man Design, Inc. v. PAJ, Inc., 262 F.3d 101, 110 (2d Cir.2001) (“Under the Copyright Act, one may market a product identical to a copyrighted work so long as the second comer designed his product independently.”); Repp v. Webber, 132 F.3d 882, 889 (2d Cir.1997); Eden Toys v. Marshall Field & Co., 675 F.2d 498, 501 (2d Cir.1982). Actual copying may be established by direct evidence, such as by defendant’s admission or by an eyewitness account. More commonly, copying is established indirectly by evidence of “access to the copyrighted work, similarities that are probative of copying between the works, and expert testimony.” Laureyssens v. Idea Group, Inc., 964 F.2d 131, 140 (2d Cir.1992). Generally, it is the conjunction of access and probative similarity that lead to an inference of direct copying. See id. (“A common form of indirect proof of copying' — but far from the only form — is a showing of defendant’s opportunity to come into contact with plaintiffs work and such similarities between the works which, under all the circumstances, make independent creation unlikely.”). “Access may be established directly or inferred from the fact that a work was widely disseminated or that a party had a reasonable possibility of viewing the prior work.” Boisson v. Banian, 273 F.3d 262, 270 (2d Cir.2001). Proof of access may also be inferred where “two works are so strikingly similar as to preclude the possibility of independent creation.... ” Lipton v. Nature Co., 71 F.3d 464, 471 (2d Cir.1995); Repp, 132 F.3d at 889 & n. 1. As the court of appeals for this circuit noted in Repp v. Webber, “copyright case-law has caused considerable confusion by the use of the term “substantial similarity” at two different points of the copyright infringement analysis.” Repp, 132 F.3d at 889. The court attempted to resolve this confusion in Laureyssens v. Idea Group, Inc., 964 F.2d 131 (2d Cir.1992), where it observed that the similarity, as it relates to proving actual copying, may or .may not be substantial. [Evidence of similarity at this stage is] not offered for [its] own sake in satisfaction of the requirement that defendant has *160 taken a substantial amount of protected material from the plaintiffs ■ work. Rather, [it is] offered as probative of the act of copying and may accordingly for the sake of clarity conveniently be called ‘probative similarity.’ Id at 140. Subsequent decisions have reintroduced uncertainty, particularly regarding whether the relevant comparison is limited to protectible elements. Compare Fisher-Price, Inc., 25 F.3d at 123, (“In the context of deciding whether the defendant copied at all (as distinguished from whether it illegally copied), ‘similarity’ relates to the entire work, not just the protectible elements.”) (emphasis in original) with Williams v. Crichton, 84 F.3d 581, 587 (2d Cir.1996) (“In the absence of direct evidence, copying is proven by showing (a) that the defendant had access to the copyrighted work and (b) the substantial similarity of protectible material in the two works”) (emphasis added). Although the Repp court did not address this issue, it reaffirmed the Laureyssens court’s distinction between “probative” similarity, for the purpose of proving actual copying, and “substantial” similarity, for the purpose of showing infringement. Repp, 132 F.3d at 889 & n. 1. An appropriate interpretation of the rule is that an examination of probative similarity is not confined to aspects of the work which are protected against infringement;, any similarity supporting an inference of actual copying is properly considered in deciding that copying occurred. b. Illegal Copying i. Protectible Elements The court must determine which, if any, of the elements are protecti-ble before proceeding to compare the works. A certificate of registration establishes a presumption that the work and its elements are original and protectible. Boisson v. Banian, Ltd., 273 F.3d at 268. Nevertheless, “[s]imply because a work is copyrighted does not mean every element of that work is protected.... [A]n element within a work may be unprotectible even if other elements, or the work as a whole, warrant protection.” Id. A variety of reasons lead to an element being characterized as unprotecti-ble. For one, copyright protects particular expressions of an idea, but not the idea itself. Differentiating between an idea and its expression is inherently imprecise. As Judge Learned Hand recognized: Upon any work, ... a great number of patterns of increasing generality will fit equally well, as more and more of the incident is left out. The last may perhaps be no more than the most general statement of what the [work] is about, and at times might consist only of its title; but there is a point in this series of abstractions where they are no longer protected, since otherwise the [author] could prevent the use of his ‘ideas,’ to which, apart from their expression, his property is never extended. Nichols v. Universal Pictures Corp., 45 F.2d 119, 121 (2d Cir.1930), cert. denied, 282 U.S. 902, 51 S.Ct. 216, 75 L.Ed. 795 (1931). An example of this rule in practice is the unprotectibility of “scenes a faire”, which are similarities of incident or plot or scenes that “necessarily result from the choice of a setting or situation.” Walker v. Time Life Films, Inc., 784 F.2d 44, 50 (2d Cir.1986); see also 4 Nimmer § 13.03[B][4] at 13-74. In Mattel, Inc. v. Azrak-Hamway Intern., Inc., 724 F.2d 357 (2d Cir.1983), a case involving action figurines, the concept of a “superhuman muscleman crouching in what since Neanderthal times has been a traditional fighting pose” constituted an unprotectible idea, but the *161 choices by the artists of how precisely to exaggerate the anatomy were protectible. Id. at 360. Elements which are primarily utilitarian are also not protectible. “Useful articles”, which are defined as “article[s] having an intrinsic utilitarian function that is not merely to portray the appearance of the article or to convey information”, 17 U.S.C. § 101, are not themselves copyrightable. Knitwaves, Inc. v. Lollytogs, Ltd., 71 F.3d 996, 1002 (2d Cir.1995) (clothing not copyrightable). Artistic aspects of utilitarian articles that are conceptually separable from the useful work may be copyrighted. See Kieselstein-Cord v. Accessories by Pearl, Inc. 632 F.2d 989, 996 (2d Cir.1980) (majority opinion; dissenting opinion found difficulty in disentangling' the artistic and utilitarian aspects of the article in question). Finally, elements which are not original are unprotectible. An important consequence of this rule is that elements copied from those in the public domain are not copyrightable. A defendant alleging that parts of plaintiffs work were copied from what was in the public domain bears the burden of proving that allegation. Boisson, 273 F.3d at 269. ii. Substantial Similarity The general test for determining whether two works are substantially similar for the purposes of infringement is that of the “ordinary observer.” This “ordinary observer” is sometimes referred to as an “average lay observer,” Yurman, 262 F.3d at 111 (citations omitted), or a “reasonable observer.” See Williams v. Crichton, 84 F.3d at 589-90. The trier is to ask whether “the ordinary observer, unless he set out to detect the disparities, would be disposed to overlook them, and regard their aesthetic appeal as the same.” Hamil America, Inc., v. GFI, 193 F.3d 92, 100 (2d Cir.1999) (quoting Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487, 489 (2d Cir.1960) (L. Hand, J.)). “If an average lay observer would recognize the alleged copy as having been appropriated from the copyrighted work, then the two products are substantially similar.” Yurman, 262 F.3d at 111 (quotation marks and citations omitted). This is an objective criterion. It invokes some supposedly general sensibility, rather than the perceptions of any particular audience. The district judge hearing the case or the appellate judge reviewing it is considered competent to play the role of an “ordinary” observer. See Boisson, 273 F.3d at 272 (“We review de novo the district court’s determination with respect to substantial similarity because credibility is not at stake and all that is required is a visual comparison of the products — a task we may perform as well as the district court.”). Presumably the judge will put aside any special sensibilities, aesthetic idiosyncrasies, or observational skills when clothed with the hoi polloi role. The essentially objective test applied in the Second Circuit can be contrasted with the more subjective approach in the Fourth Circuit, where substantial similarity is judged by whether works “express those ideas in a substantially similar manner from the perspective of the intended audience of, the work.” Lyons Partnership L.P. v. Morris Costumes, Inc. 243 F.3d 789, 801 (4th Cir.2001) (citation omitted and emphasis added). Slight reverberations in Second Circuit case law suggest that the perceptions of the specialized intended audience may be significant, see Repp, 132 F.3d at 889, quoting Arnstein v. Porter 154 F.2d 464, 473 (2d Cir.1946) (“defendant took from plaintiffs works so *162 much of what is pleasing to the ears of lay-listeners, who comprise the audience for whom such ... music is composed ”) (emphasis added), but this subjective paradigm has not taken root in the Second Circuit. See, e.g., Williams v. Crichton, 84 F.3d 581, 589, 590 (examining similarity between children’s books as it would appear to the reasonable lay observer, and noting that “copyright law is to be uniformly applied across a variety of media and audiences”). While no mention of the child’s perspective in determining similarity between dolls is found in the cases, some juvenilization of a judge’s appreciation of what is important in dolls is tolerable. Where a work consists of protectible and unprotectible elements, a somewhat refined version of the “ordinary observer” test has sometimes been applied. In Folio Impressions, Inc. v. Byer California, 937 F.2d 759 (2d Cir.1991), the court was troubled that “the ordinary observer would compare the finished product that the fabric designs were intended to grace (women’s dresses), and would be inclined to view the entire dress — consisting of pro-tectible and unprotectible elements—as one whole.” Folio Impressions, 937 F.2d at 765. To avoid finding infringement based in part on similarities of unprotecti-ble features, the court concluded that the “observer’s inspection must be more discerning.” Id. at 766. In Boisson v. Banian, Ltd., 273 F.3d 262 (2d Cir.2001), the court explored how the “more discerning” test should be applied, and concluded that it would be improper to “dissect the works at issue into separate components and compare only the copyrightable elements.” Boisson, 273 F.3d at 272. Instead, the court reasoned that it should conduct an examination of the “ ‘total concept and feel’ of ... works ... instructed by common sense.” Id. at 273 (citations omitted). The Boisson opinion did not elaborate on how the “total concept and feel” accounts for unprotectible elements. An inherent ambiguity in the “total concept and feel” test hinders its rigorous explication. Cf. 4 Nimmer § 13.03[A][l][c] at 13-39 (criticizing the “total concept and feel” standard and noting that ‘feel’ is a “wholly amorphous referent” which “invites abdication of analysis”). The Boisson court’s lengthy discussion of how to define the “more discerning” test suggests that there must be some accounting for unprotectible elements; otherwise, the Boisson court would have saved time and avoided confusion by explicitly collapsing the “ordinary” and “more discerning” tests and rejecting the division created in Folio Impressions. Boisson seems best read as meaning that the discerning observer should distinguish between protectible and unprotecti-ble elements, put the unprotectible elements out of mind, and determine whether the remainders of each work, taken together, are similar in total concept and feel. See 4 Nimmer, § 13.03[E][2] at 13-84 (“the comparison should take place after filtering out of the analysis elements of plaintiffs work that are not protectible”); Hamil America, 193 F.3d at 101 (the “more discerning” ordinary observer test “requires the court to eliminate the unpro-tectible elements from its consideration and to ask whether the protectible elements, standing alone, are substantially similar”). As the obverse of similarity, dissimilarity occasionally bears on the infringement analysis. Generally, the “key to the ‘ordinary observer’ test is ... the similarities rather than the differences.” Durham Indus., 630 F.2d at 913 (citation omitted). Thus, “slight differences between a protected work and an accused work will *163 not preclude a finding of infringement.” Id. Learned Hand put it this way in Sheldon v. Metro-Goldwyn Pictures Corp., 81 F.2d 49, 56 (2d Cir.1936): “it is enough that substantial parts were lifted; no plagiarist can excuse the wrong by showing how much of his work he did not pirate.” There is some point at which changes, not in themselves sufficient to make a difference, together become substantial enough. The expression of the accused work is then no longer the same as that of the protected work, although the fundamental incorporation of the idea may have been copied. “Even if an- alleged copy is based on a copyrighted work, ‘a defendant may legitimately avoid infringement by intentionally making sufficient changes in a work which would otherwise be regarded as substantially similar to that of the plaintiffs.’ ” Eden Toys v. Marshall Field & Co., 675 F.2d 498 (2d Cir.1982), quoting Warner Bros. v. Am. Broadcasting Cos., 654 F.2d 204, 210; 4 Nimmer § 13.03[B][1][b] at 13-56. Differences become particularly important where copyright protection is “thin.” Commonly used to refer to compilations of facts in the public domain, “thin” copyright protection is appropriate where works reflect “scant creativity.” 4 Nimmer § 13.03[A] at 13-28. Scantiness may exist because the work is composed of elements in the public domain, and it is only the organization of those elements that is protectible. Feist Publ’ns. Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 348, 111 S.Ct. 1282, 1296, 113 L.Ed.2d 358 (1991). Creativity also may be scant because the author has chosen to express an idea without much embellishment, so that the expression is nearly indistinguishable from the idea itself. See Harper & Row, Publishers, Inc. v. Nation Enters., 471 U.S. 539, 105 S.Ct. 2218, 85 L.Ed.2d 588 (1985) (“[Even] within the field of fact works, there are gradations as to the relative proportion of fact and fancy.' One may move from sparsely embellished maps and directories to elegantly written biography. The extent to which one must permit expressive language to be copied, in order to assure dissemination of the underlying facts, will thus vary from case to case.”); Eng’g Dynamics, Inc., v. Structural Software, Inc., 26 F.3d 1335 (5th Cir.1994) (“[T]he scope of protection afforded by a copyright is not constant across all literary works.... [T]he law is more protective of highly original and highly expressive works than it is of functional and nonfiction works.”). This concept of thinness is separate from, but related to, the doctrine of merger, which was recently summarized: “if there is just one way to express an idea, the idea and expression are said to merge, and the expression is not protecti-ble. The test is whether protection of expression would inevitably accord protection to an idea.” Yurman, 262 F.3d at 111-12 (quotation, quotation marks and citations omitted). It follows that, where an author expresses an idea at a level specific enough to avoid denial of copyright under the merger doctrine, but general enough that considerable similarity will arise in any other expression of the idea, the author’s copyright is not defeated by the doctrine of merger. For this category of cases, however, other authors must be permitted to escape liability for infringement by relying upon the differences in their work, to avoid having the original author, for all practical purposes, appropriate the idea itself. So in Mattel, Inc. v. Azrak-Hamway Intern., Inc., where the court held that although “the dolls’ bodies are very similar, nearly all of the similarity can be attributed to the fact that both are artist’s rendering of the same unprotected idea,” the fact that “a lay observer would *164 recognize certain differences” in the expression of the ideas was sufficient to avoid a finding' of substantial similarity. Mattel, Inc. v. Azrak-Hamway Intern., Inc., 724 F.2d at 360. Even where a work merits only “thin” protection, de minimis changes (such as adding a single .entry to an otherwise identical copy of a protected Phonebook) will not reheve the infringer of liability. See Key Publications, Inc. v. Chinatown Today Pub. Enters., Inc., 945 F.2d 509, 514 (2d Cir.1991) (thin does not mean anorexic). V. Application of Law to Facts A. Registration There is no dispute that plaintiff satisfied the registration requirement with regards to the 20 inch Sweetie Mine doll. 17 U.S.C. § 411(a) There is also no question that plaintiffs have not registered a copyright in the 48 inch Sweetie Mine doll, and have never attempted to do so. Plaintiffs infringement claim is rooted solely in the registered copyright of the 20 inch doll. It contends that no separate registration of the 48 inch doll was necessary because the copyright is not size-specific. It calls attention to the fact that its 48 inch doll was created by using a photocopy machine to enlarge the fabric patterns used for the 20 inch doll, while admitting that design adjustments were made for the larger doll. It argues that these changes were solely intended to ensure that the 48 inch doll was “true to” the original 20 inch doll. Thus, plaintiff claims, its 48 inch doll includes no original contributions beyond those present in the 20 inch doll, so that both are equally protected under the registered copyright. Defendant counters that the plaintiffs 48 inch doll was designed with different proportions than the 20 incher, and that the changes amounted to substantial original artistic input, requiring a separate registration before any federal suit is heard on infringement of the larger toy’s copyright. Plaintiffs copyright in its 48 inch doll is not subsumed in the registration of the 20 inch doll. The 48 inch figure is not an exact replica of the 20 inch figurine, nor is it a scale model. Important changes were made in the proportions of the various parts and their relation to one another. For example the larger doll’s head size was made smaller relative to the rest of the body and juts our more from the body; and the legs are longer and thinner. The facial features on the 48 inch doll are more spread out and higher on the face compared to those of the 20 inch doll. The braids of the 48 inch doll are shorter relative to the body of the doll than are the braids of the 20 inch doll. The shoulders of the 48 inch manikin appear more pronounced than those of the 20 inch model. All of the witnesses agreed thafchanges in proportion were necessary to make the plaintiffs 48 inch Sweetie-Mine doll appear pleasing. The problems in scaling up dolls is analogous to those that would be involved in blowing up a 6x8 inch picture of an infant to a 6 foot poster — the proportions would not suggest an adult. The parties concede that there were a range of options for the designers. Artistic discretion, rather than some mathematical formula or mechanical convention, guided their choices. The changes actually made by plaintiff in increasing the size of its 20 incher were not trivial. They produced a different “aesthetic appeal” in the 48 inch plaything than existed in the smaller one. Since some of the substantial differences between the 48 and 20 inch dolls are the product of “artistic” rather than “mechanical” skill, under Durham and L. Batlin & Son, Inc. the 48 inch doll is the proper *165 subject of a derivative copyright. Because, however, the plaintiffs 48 inch doll is unregistered, the court lacks subject matter jurisdiction to hear any claims arising from its infringement. Plaintiff is correct, of course, that the failure to register the derivative 48 inch doll in no way affects plaintiffs right to sue on the basis of the 20 inch version. The derivative copyright in the 48 incher, if it had been registered, would protect only those elements original to the 48 inch doll. Elements imitating those found in the 20 inch doll would be protected under the registered copyright in the 20 inch doll. To whatever extent Well-Made’s 48 inch doll incorporates protectible elements of its 20 inch doll, the infringement of those elements by an unauthorized copy of the 48 incher would be actionable. B. Infringement 1. Valid Copyright The first element of the Feist test, ownership of a valid copyright, is satisfied here. Plaintiffs registered copyright in the 20 inch Sweetie Mine doll, issued in April, 1996, is presumptively valid. Defendants do not contest validity. 2. Copying of Original Elements a. Actual Copying As previously pointed out, the plaintiff must prove that the copyright in the 20 inch doll was infringed. There is no evidence to suggest that defendant directly copied the design of plaintiffs 20 inch doll. There is, however, persuasive evidence that defendant copied plaintiffs 48 inch doll. This raises the question: in order to prove infringement of the copyright in the 20 inch doll, must plaintiff prove that the 20 inch doll was itself actually copied, or would copying the 48 inch doll suffice? The answer is that evidence of actual copying of the 48 inch doll suffices. There is a transitive property to actual copying: if work A is an actual copy of work B, and work B is an actual copy of work C, then work A is deemed an actual copy of work C. Pye v. Mitchell, 574 F.2d 476 (9th Cir.1978); De Acosta v. Brown, 146 F.2d 408, 411-12 (2d Cir.1944); 3 Nimmer § 8.01[C] at 8-18 (“[I]t is no defense even if the defendant did copy from a third work rather than from the plaintiff if such third work was itself an unauthorized copy of the plaintiffs work”). In the words of Judge Learned Hand, “when one copies a copy, he copies the original.” De Acosta v. Brown, 146 F.2d 408, 413 (2d Cir.1944) (L. Hand, J., concurring). This rule does not apply merely to exact replicas; it also applies to derivatives. Cf. De Acosta v. Brown, 146 F.2d 408 (magazine publishing extracts from unpublished book which was a .plagiarism of an uncopyrighted screen play liable for plagiarism). To paraphrase Learned Hand, when one copies a derivative, he copies the original. It changes nothing that the .derivative and the original share the same author. To hold otherwise would violate, at least in spirit, the rule that the “scope of protection afforded a derivative work ... must not in any way affect the scope of any copyright protection in that preexisting material.” Durham, 630 F.2d at 909. To prove access, plaintiff need only show that the work was “widely disseminated” or that there was a “reasonable possibility” that defendant could have examined plaintiffs work; proof that defendant actually viewed it is unnecessary. Boisson, 273 F.3d at 270. Both standards are met here. Goffa’s president, Mr. Song, testified that he decided to design and market a 48 inch rag doll only after *166 learning that there was a 48 inch rag doll selling well in the United States. Goffa learned this from Well-Made’s former distributor. Goffa in fact did obtain a sample of Well-Made’s 48 inch doll by purchasing one. There is also the likelihood that Hu-bei in China made its 48 inch sample-— furnished to defendant—from a modified copy of plaintiffs 20 inch doll. Access alone does not prove actual copying. There must be probative similarity between the protected work and the allegedly infringing work. When comparing works for the purpose of determining probative similarity, protectible and unprotectible elements need not be differentiated. See supra III.B.2.a. The court is not required to view the work in totality; dissection is permitted. See Walker v. Time Life Films, 784 F.2d 44, 51 (2d Cir.1986). To demonstrate probative similarity, plaintiffs disassembled into its fabric pieces defendant’s 48 inch doll and superimposed them upon the manufacturing patterns of Well-Made’s 48 inch doll. Defendant contends that this evidence is unreliable because: 1) the process used to mass produce fabric pieces is inherently imprecise, causing them to vary from the templates; and 2) after being sewn together and stuffed, the pieces are stretched and distorted. The believable testimony on this matter suggested that any distortions were likely to have been slight. Defendants had the opportunity to introduce their manufacturing patterns or more accurate fabric and paper pattern samples if they believed the samples in evidence to be inaccurate; they failed to do so. The court finds that the fabric pieces and paper patterns introduced by plaintiff are reliable for the purpose of establishing probative similarity. Defendant’s fabric pieces do not match plaintiffs manufacturing patterns exactly. Most are quite close, however, differing by less than an inch in any dimension. The notable exceptions are the sleeve, for which Goffa’s fabric piece is approximately two inches longer than the Well-Made pattern piece, and the skirt, for which Goffa’s piece is approximately three inches longer and two inches wider than Well-Made’s pattern. The Goffa legs seem somewhat wider and the face has smaller tucks, leading to a flatter visage. Defendants incorrectly argue that dissimilarities in the facial embroidery, fabric patterns, and color choices preclude a finding of actual copying. These dissimilarities only demonstrate that, as to those elements, there has been no copying. They do not bear on whether the fabric pieces comprising the body sculpture were copied. Similarity between any one or two fabric pieces might be happenstance. The close similarity between all but a few of the pieces, however, reduces almost to the vanishing point any likelihood that the similarities are coincidental. Nonetheless, Goffa claims that its doll was the product of independent creation. As evidence, it offered several faxes and photographs purporting to be documents contemporaneous with the design of Gof-fa’s doll. This evidence is not reliable. It does not prove independent creation. Gof-fa claims that it instructed its factory to design a 48 inch doll amalgamating features from several previous Goffa dolls. It is clear that the factory did copy the face for the new doll from a preexisting Goffa doll, “No. YL 11352”. It is also possible that the arm design was copied from a preexisting Goffa doll, “No. SL 11852”. But the remainder of the shape and the body of the 48 inch doll, which were allegedly copied from a preexisting Goffa doll, “No. SL122”, bear little resemblance to *167 that doll, or to any other earlier Goffa product. Whether or not Goffa actually instructed its factory to copy the body design for the 48 inch doll from SL122, it is obvious the factory did not do so. The evidence demonstrates that defendant’s factory must have largely copied its 48 inch design from somewhere, since defendant’s own witnesses testified that the factory had never produced ragdolls before, was “not creative” when it came to product design, and that its employees “couldn’t do much on their own.” Tellingly, Goffa produced no witnesses from the Shanghai factory to testify about the design process. In addition to the usual indirect means of proving copying, in this case actual copying was directly demonstrated. The court finds by a preponderance of the evidence that defendant actually purchased and examined plaintiffs 48 inch work and had in its possession the Hubei 48 inch version based upon a copy of plaintiffs 20 inch doll before it designed its own 48 inch doll. Both of those templates were furnished by defendant to its factory in China which copied substantial portions of one or both of them in producing the final design for defendant’s 48 inch doll. It is clear beyond reasonable dispute, based on the evidence, that one of the dolls copied by defendant’s factory was the 48 inch doll of plaintiff that had been purchased by defendant at retail. Based on the defendant’s access to plaintiffs doll, the striking similarity between many of the fabric pieces, and defendant’s failure to provide credible evidence of independent creation, the court finds that defendant actually copied large parts of plaintiffs 48 inch Sweetie Mine doll, and, indirectly, some of the expanded parts of the 20 inch doll as well. b. Illegal Copying. Although plaintiffs may satisfy the actual copying prong by showing that Gof-fa actually copied significant parts of Well-Made’s 48 inch doll, the illegal copying prong may only be satisfied by reference to the registered work, Well-Made’s 20 inch doll. Plaintiffs have only claimed infringement in the 20 inch copyright, and thus must prove substantial similarity between Goffa’s 48 inch doll and the protecti-ble elements of Well-Made’s 20 inch doll. Fisher-Price, Inc. v. Well-Made Toy Manufacturing, 25 F.3d 119, 122-23 (2d Cir.1994). Comparisons to Well-Made’s plaintiffs 48 inch doll would only confuse the analysis by introducing elements that are original to the 48 inch work and thus unprotectible in this suit based on the registered copyright in the 20 inch doll. Defendant’s 48 inch doll does not infringe' plaintiffs 20 inch doll. Even if plaintiffs own 48 inch doll does not depart so substantially from its own 20 incher, as to miss being distinctive, defendant’s 48 incher makes further changes from plaintiffs so that it differs substantially from plaintiffs 20 inch doll. Not having registered its 48 inch doll; plaintiff cannot sue in this court for a resemblance to that doll. The first step in analyzing the dolls for substantial similarity is .to determine whether any elements of the 20 inch doll are unprotectible. Major aspects of the doll are stipulated to be intrinsic to the idea of ragdolls. These elements are un-protectible because they constitute part of the “idea” of a ragdoll. They include such aspects as shaped heads, torsos, embroidered faces, hats, bows, yarn hair, ribbons, dresses, hands, feet, and shoes. Any similarity arising from the fact that both toys possess some form of these intrinsic features is non-infringing similarity. Connie Rossi, who teaches soft toy design at the Fashion Institute in New York City, and whose testimony for the plaintiff the court finds credible, testified that *168 many of the body sculpture components of these dolls were basic and “commonly used pattern pieces.” This comports with the court’s observation of the dolls; the legs are cylinders, with almost shapeless bulbs for feet; the skirt and bodice have simple contours; the modeling of the arms and head, are more distinctive but still quite uniform and relatively gross in form, lacking any fine defining features. None of the general rag doll technology renders these shapes unprotectible; neither party has claimed, nor has the evidence shown, that any elements of plaintiffs design were copied from the public domain. The uncreative, basic nature of these shapes, and the fact that similar shapes will necessarily be present in any ragdoll mean, however, that plaintiffs sculptural designs will receive only narrow protection; otherwise, plaintiffs would perversely be rewarded for their lack of creativity with, effectively, a copyright in the idea of all ragdolls. The “pocket” or “cell” design of the face is not a protectible feature, since this is a widely used manufacturing technique of utilitarian value. See Little Souls, Inc. v. Petits, 789 F.Supp. 56, 58 (D.Mass.1992) (citations omitted). Ms. Rossi testified that she teaches her students the technique of using such cells to achieve appealing looks in doll-like stuffed animals. The test for substantial similarity must be taken from the perspective of the “ordinary observer.” Hamil America, 193 F.3d at 100. Yet an ordinary 80 year old judge is bound to see a 48 inch rag doll somewhat differently than an ordinary 8 year old child. If the court perceived, as did one witness whom might fairly be described as an ordinary middle aged manager, that “all dolls look the same,” copyright protection would extend too far. No young children testified, but one need not speculate unduly to conclude that the ordinary youngster would name each of the dolls, see in their faces and postures very different personalities, and have no difficulty at all in pointing to differences he or she perceived to be significant. Or, alternatively, one could imagine the child being even less discerning than adults. Cf. Lyons Partnership L.P. v. Morris Costumes, Inc., 243 F.3d 789, 801-03 (4th Cir.2001) (adults could distinguish between dinosaur costumes but intended audience of 2-5 year old children could not). It is impossible to know without asking, and that was not done here because neither the parties nor the law required the inquiry. Mindful that the court must evaluate the similarity in “total concept and feel” of the plaintiffs doll and defendant’s doll, it makes the following findings: the 20 inch Sweetie-Mine doll has a considerably younger, more winsome look than the defendant’s 48 inch doll; each possesses a different “aesthetic appeal.” This is in part due to the obvious difference in size. Different body proportions and facial features also influence this impression, as do different fabric and color choices. What similarities exist arise essentially from both dolls being expressions of the idea of a female rag doll. Most apparent is the difference in size. Goffa’s 48 inch doll is 2.4 times the size of Well-Made’s copyrighted 20 inch doll. Scale, or how the experience of a sculptural work relates to the viewer’s body, plays an important role in defining the aesthetic appeal of artistic works. See Lilian Tone, Size Specific, in Modern Contemporary: Art at MoMA Since 1980, at 534 (Kirk Yarnedoe, Paola Antonelli & Joshua Seigel eds., 2000). Robert Morris has explained, In the perception of relative size, the human body enters into the total continuum of sizes and establishes itself as a constant on that scale. One knows im *169 mediately what is smaller and what is larger than himself. It is obvious yet important to take note of the fact that things smaller than ourselves are seen differently than things larger. The quality of intimacy is attached to an object in a fairly direct proportion as its size diminishes in relation to oneself. Robert Morris, Notes on Sculpture, Part 2, in Continuous Project Altered Daily: The Writings of Robert Morris 11 (1993). The court agrees with Ms. Speer’s observation that the plaintiffs 48 inch doll “has a different relationship to a person” than the 20 inch version. Objects as ordinary as a safety pin and garden trowel can take on artistic significance worthy of a special exhibition at the Metropolitan Museum of Art in New York City — as long as the pin and trowel are twenty-one and twenty four feet high, respectively. See Howard Kissel, Met Roof Gets a Rare Case of Claes, Daily News, May 10, 2002, at 63 (works of Claes Oldenburg and Coosje van Bruggen exhibited at the Metropolitan Museum of Art). Size matters. To a young child, the difference between a 48 inch doll and a 20 inch doll would be particularly powerful— the 48 inch doll would be as tall, or perhaps taller than the child, while the 20 inch doll would be much shorter than she or he. To a full-grown adult, the difference is less powerful, but is striking nonetheless. The 48 inch doll conveys the impression of a person; during the course of this trial, the 48 inch dolls were sat in the chairs normally reserved for the jury and jokingly described as jurors by counsel. The 20 inch doll appears to be an infant, not to be mistaken for an older child or an adult. Nevertheless, if the proportions vof the 48 inch doll were identical to those of the 20 inch work from which it was derived, as those of the Oldenberg sculptures displayed at the Metropolitan Museum of Art appear to be, and assuming no other changes, the copyright (and registration) of the smaller could be assumed to incorporate the larger, at least in the case of dolls. But the increase in size was not based on a mechanical increase in scale, but on aesthetic .choices. Connected to the difference in size, but conceptually separate, are the differences in the relative proportion of body parts. Consider that, in a photograph, a six foot tall man can be reduced to a six inch image, and the reduction is uniform in scale; yet the image is instantly recognizable as a- man and need not be altered to appear veracious. Leonardo da Vinci’s famous sketch of a nude man with arms and legs outstretched, the “Vitruvian Man,” attests to the fact that adult human form adheres, more or less, to certain proportions. Da Vinci’s sketch was inspired by a passage from Vitruvius’s De Architectural: For Nature has so planned the human body that the face from the chin to the top of the forehead and the roots of the hair is a tenth part; also the palm of the hand from the wrist to the top of the middle finger is as much; the head from the chin to the crown, an eighth part; from the top of the breast with the bottom of the neck to the roots of the hair, a sixth part; from the middle of the breast to the crown, a fourth part; a third part of the height of the face is from the bottom of the chin to the bottom of the nostrils; the nose from the bottom of the nostrils to the line between the brows, as much; ■ from that line to the roots of the hair, the forehead is given as the third part. The foot is a sixth of the height of the body; the cubit a quarter, the breast also a quarter. The other limbs also have their own proportionate measurements. Viruvius Pollio, De Architectural Book 3, c. I(Trans.F.Granger, 1962). *170 Vitruvius and da Vinci were examining the adult form. Our proportions do not remain constant throughout our lives. A human baby has a large head and short, thick limbs relative to its torso; its facial features are set close together and low on the face (creating a high forehead). As the child grows to adulthood, the limbs lengthen and thin out, the body grows more quickly than the head. See, e.g., Ronald W. Mayer, Module 4: The Beginnings of Behavior — Physical Growth and Behavior, at http://online.sfsu.edu/'— psych200/unit2/24.htm (“[I]t is apparent that a child’s entire skeletal structure is proportionally different from that of an adult.... The different parts.of the body do not grow at the same rate, or at the same time. Instead,. the head develops first, the torso next, and the limbs last.... At birth a baby’s head is about 25% of the total body length, or about 60% of its adult size An adults head is normally less than 15% of the total body length”); Drawing and Painting Life Portraits: Drawing the Infant Head in Profile, at http://www.wet-canvas.com/Art School/Portraiture/StillLi-fePortraits/Lesson2/. In short, Goffa’s 48 inch doll is, not the expanded twin of Well-Made’s 20 inch doll. Their “DNA” is different. She is more like the 20 inch doll’s older sister or first cousin. The 20 inch doll has proportions appropriate to a human infant. If the 48 inch doll had been given the same proportions as the 20 inch doll, it would have appeared “klutzy” only because it would have seemed an unnaturally large infant. Goffa’s 48 inch doll, like Well-Made’s “grown-up” Sweetie-Mine doll, has different proportions because both are intended to have, and do have, an older appearance suitable to their size. Differences between the faces of the dolls distinguish them further. Humans have evolved the capacity to differentiate between even quite similar faces; as social animals, we depend on our ability to recognize faces as our main means of identifying those around us. See A.W. Young, Finding the Mind’s Construction in the Face, in Face and Mind 1, 53 (Andrew W. Young ed., 1998) (“The ease with which we can determine familiarity may be relatively great for faces ... because the ability is of such ecological importance.”). In general, our perceptive faculties are less sensitive to non-facial aspects of appearance. See H.D. Ellis and A.W. Young, Faces in Their Social and Biological Context, in Face and Mind 67, 67-68 (Andrew W. Young ed., 1998) (“[faces] have been described as the single most important visual pattern in our environment, and our ability to discriminate hundreds of faces ... represents the ultimate in our perceptual classification skills”) (citations omitted). As dolls are iconic of human beings, the ordinary observer is particularly attuned to their faces; he or she will likely look at the face first, and rely upon it in forming an overall impression of the doll. See id. (in face perception “one small discordance over-weighs a multitude of similarities and suggests a general unlikeness.”) (quoting F. Galton, Inquiries into Human Faculty and Development (1883)). Courts have recognized that the similarity or dissimilarity of doll faces play a key role in determining whether infringement exists, even when the toy does not imitate a human. In Fisher-Price, Inc. v. Well-Made Toy Manufacturing Corp., 25 F.3d 119 (2d Cir.1994), the court of appeals for the Second Circuit found that a mouse doll did not infringe Fisher-Price’s copyright in its mouse doll because their faces were substantially distinct. Fisher-Price, Inc. v. Well-Made Toy Manufacturing Corp., 25 F.3d at 124 (‘While these dolls have similar body types, the artistic work on the faces is entirely distinct.”). The same Fisher-Price court found that a human *171 doll did infringe on a copyrighted human doll because of their similar faces. Id. (“Both sport the same bright, painted eyes, the same skyward gaze, the same knobby nose, and the same cherubic smile.... These dolls do not merely share features that are common to all dolls; they contain virtually identical expressions of those features.”); see also Russ Berrie & Co. v. Jerry Elsner Co., 482 F.Supp. 980, 986 (S.D.N.Y.1980) (Although Santa dolls had similar body forms, the dolls had “strikingly” different facial features: “Eisner’s is a wide-eyed, almost startled looking Santa; Berrie’s a sleepy-eyed more whimsical fellow. The average observer would note the difference immediately.”). The faces of both plaintiffs and defendant’s dolls show greater detail than any other of their parts (with the exception, perhaps, of the fabric patterns — which themselves are quite different). The faces are the only parts that are embroidered and plaintiff does not allege that the face of defendant’s doll infringes on plaintiffs registered copyright. To whatever extent the faces are similar (and there are some similarities, such as the flat oval noses), defendants have proven independent creation because they designed and copyrighted the face for their CL doll, from which they copied their 48 inch doll’s face, before -plaintiff designed the face of its “Sweetie Mine.” The faces are different in obvious and important respects. Plaintiffs 20 inch Sweetie Mine doll, as the name would suggest, is a sugary faced doll, with emphatically pretty features including detailed, well-defined eyes, thick eyelashes, and a full, well-defined pink mouth with pursed lips. Prominent pink blush marks at the lower corners of the face and a high forehead give the Sweetie-Mine doll a very young look. The doll’s gaze appears to be level and directed at the viewer, creating the impression that the doll has, or desires, some connection with the viewer. The doll’s expression is one of pleasant satisfaction In contrast, defendant’s Huggable Lovable doll has a less sophisticated look. Its face is broader and its facial features drawn in less detail. The eyes do not have defined pupils and irises and, compared to the Sweetie Mine doll, the eyelashes are more sparse and located differently. Unlike the Sweetie-Mine doll, the lips are thin and not well defined, and the mouth is open. The eyes are raised upwards so that the gaze does not fall on the viewer. The doll’s expression is one of joy. Difference in color and fabric patterns also contribute to each’s,distinct feel. Similarities in color choice clearly may affect the “total concept and feel” of a work for purposes of copyright infringement. See Boisson v. Banian, 273 F.3d 262, 271, 272-76 (2d Cir.2001) (holding that “the author’s choice- in incorporating color with other elements may be copyrighted” and considering color choices when comparing works.); Streetwise Maps, Inc. v. Vandam, Inc., 159 F.3d 739, 748 (comparing the “overall manner in which [plaintiff] selected, coordinated, and arranged the expressive elements in its map, including col- or.... ”). Plaintiffs doll has blonde yarn hair and pink bows, belt and feet, enhancing her youthful look. Defendant’s doll has brown yarn hair and deep red bows, belt, and feet, giving her an older look. The fabric of plaintiffs dress is a somewhat complicated patchwork of hearts, butterflies, and geometric designs presented in a wide range of pastel colors, with pastel blue as the dominant color. The fabric of defendant’s dress is darker — a simple, traditional flower print with only three colors, red, white and blue, with red the dominant color. Plaintiffs doll’s dress suggests she is a child of the city in *172 Spring, while defendant’s doll appears to be the offspring of the countryside in Fall. There are other differences as well. Plaintiffs 20 inch doll has a bow on the back of its waist sash; defendant’s doll does not. Adjusted for the difference in size, defendant’s doll’s dress is appreciably longer than that of plaintiffs doll, which affects how much of the legs show. The upper sleeves of the arms of defendant’s doll are noticeably longer than those of plaintiffs doll, and the band at the wrist is made from the dress fabric on the plaintiffs doll, but from the sash and bow fabric on defendant’s. The defendant’s doll has a wider neck than does the plaintiffs, giving the defendant’s doll more pronounced shoulders. While none of these differences are significant on their own, in aggregate they affect the “total concept and feel” of both dolls. This case is similar to Ideal Toy Corp. v. Fab-Lu Limited, 266 F.Supp. 755 (S.D.N.Y.1964), another litigation involving dolls, where the district court held that “[although the accused dolls are similar to (appellant’s) dolls in size and shape, and indeed some features (such as hands and arms) are virtually identical, we think the total effect of the image conveyed to an ordinary observer by the accused dolls is quite distinct from that of (appellant’s) dolls.” Id. at 756. On appeal, the court of appeals affirmed, 360 F.2d 1021 (2d Cir.1966), and noted that while similarities existed as to “standard doll features such as the full faces; pert, upturned noses; bow lips; large, widely spaced eyes; and slim figures, distinct differences exist with respect to [other features].” Id. at 1023. VI. Conclusion The case here is not a simple matter of the art of gigantism, where, as in the case of the present Oldenburg exhibit, a trowel designed to be held in the hand and less than half a cubit in length is transposed, with its original proportions presumably intact, into a towering artifact that dwarfs its human admirers. In the instant case the proportions of the diminutive original and the large derivative are significantly different, the result of artistic judgement of a skilled designer. Registering the small version is not the equivalent of a registration of the significantly larger and different rendering. Had the suit been premised upon infringement of plaintiffs copyright in its 48 inch doll, the suit would have been dismissed for lack of jurisdiction because there was no registration for the larger doll. Plaintiff sought to avoid this result by suing for infringement of its 20 inch doll. But the differences between plaintiffs 20 inch doll and defendant’s 48 inch doll are too great. “[Wjhatever unique quality [plaintiffs 20 inch] doll may have, [defendant has] not appropriated it.” Ideal Toy Corp. v. Fab-Lu Limited, 360 F.2d at 1023, quoting Ideal Toy Corp. v. Sayco Doll Corp., 302 F.2d 623, 627 (2d Cir.1962) (Clark, J., dissenting). There is no substantial similarity for the purpose of infringement. Defendant’s 48 inch doll does not infringe plaintiffs registered 20 doll. Enter judgement for the defendant. No costs or disbursements. SO ORDERED.