高分辨率 CT 掃描數(shù)據(jù)揭示晚侏羅世暴龍類五彩冠龍牙齒替換模式

High-resolution CT-scan data reveals the tooth replacement pattern of the Late Jurassic tyrannosauroid Guanlong wucaii （Dinosauria， Theropoda）

Abstract The Tyrannosauridae， which is characterized by specialized pachydont dentition andputative bone-cracking predatory strategies， is one of the most extensively studied theropodlineages. Although tooth replacement patterns， crucial for understanding feeding behaviors， havebeen thoroughly studied in this group， studies on non-tyrannosaurid tyrannosauroids are relativelyscarce. This study utilizes high-resolution CT data to investigate the tooth replacement pattern intwo specimens of Guanlong wucaii， a Late Jurassic tyrannosauroid， and provides insights into theevolution of tooth replacement across Tyrannosauroidea. Second-generation replacement teeth， ararity observed mainly in giant predatory theropods （e.g. some tyrannosaurids）， were detected inthe dentary dentition of the juvenile Guanlong. Zahnreihen reconstructions display a consistentcephalad alternating tooth replacement pattern in the maxilla and the dentary of both of theexamined individuals， with Z-spacing values exceeding 2.0. As Guanlong grows， the Z-spacingvalue in the maxillary dentition increases， resembling the ontogenetic changes documented in theTyrannosauridae. Additionally， like Tarbosaurus， Guanlong also displays a discontinuity betweenthe tooth replacement waves at the premaxilla-maxilla boundary. This study thus demonstratesthat some tyrannosaurid-like tooth replacement patterns were acquired before the origin of theTyrannosauridae.

Key words Late Jurassic， Tyrannosauroidea， tooth replacement， 3D reconstruction

Citation Ke Y H， Pei R， Xu X， 2024. High-resolution CT-scan data reveals the tooth replacementpattern of the Late Jurassic tyrannosauroid Guanlong wucaii （Dinosauria， Theropoda）.Vertebrata PalAsiatica， 62（3）： 225–244

1 Introduction

The Tyrannosauridae， including Tyrannosaurus rex and its close relatives， is a group ofcarnivorous theropods with a large body size and a highly specialized body-plan， and stands asone of the most extensively studied theropod groups （Osborn， 1905; Osborn and Brown， 1906;Lambe， 1917; Brochu， 2003; Currie， 2003; Hurum and Sabath， 2003; Horner and Padian， 2004;Brusatte et al.， 2010; Tsuihiji et al.， 2011; Brusatte and Carr， 2016; Carr， 2020; Woodward etal.， 2020; Marshall et al.， 2021; Paul et al.， 2022; Dalman et al.， 2023; Scherer and Voiculescu-Holvad， 2024）. As the apex predators of the Late Cretaceous， tyrannosaurids harbor highlyspecialized feeding apparatuses that have attracted lots of attention （Currie et al.， 1990; Abler，1992; Farlow and Brinkman， 1994; Erickson， 1995; Smith， 2005; Buckley et al.， 2010; Reichel，2012; Owocki et al.， 2020; Funston et al.， 2021; Holtz， 2021; Therrien et al.， 2021）. These giantcarnivorous dinosaurs， which were hypothesized to employ a unique bone-cracking strategy，exhibit some derived dental features， including extremely thickened teeth and heterodonty（refer to pseudo-heterodonty， sensu Hendrickx et al.， 2015） in both morphology and function（Erickson and Olson，1996; Erickson et al.， 1996; Hurum and Currie， 2000; Meers， 2002;Brochu， 2003; Rayfield， 2004; Smith， 2005; Hone and Watabe， 2010; Reichel， 2010; DePalmaII et al.， 2013; Gignac and Erickson， 2017; Peterson et al.， 2021; Rowe and Snively， 2022;Winkler et al.， 2022; Therrien et al.， 2023）.

In recent studies， tooth replacement patterns have been found to provide insights intothe feeding behaviors of most non-mammalian gnathostomes （D’Emic et al.， 2019; Maho etal.， 2022; Maho and Reisz， 2024）. The Tyrannosauridae was among the first theropod groupsstudied for tooth replacement， serving as an iconic model for understanding theropod dentalcycling. Lambe （1917） first described the alternating tooth loss in Gorgosaurus. Utilizing thehistological thin sectioning method， Edmund （1960） further investigated the tooth replacementpatterns of non-mammalian vertebrates including large-bodied theropods. The widespreadadoption of CT scanning technology has facilitated the non-destructive visualization ofreplacement teeth imbedded in the jaw bones， leading to more comprehensive investigationson tooth replacement patterns in theropods （Lautenschlager et al.， 2014; Dumont et al.， 2016;Kundrát et al.， 2019; D’Emic et al.， 2019; Wu et al.， 2021; Powers et al.， 2021） includingtyrannosaurids （Erickson， 1996; Hanai and Tsuihiji， 2018; Sattler and Schwarz， 2021）.Tarbosaurus and Gorgosaurus were characterized by the presence of second-generationreplacement teeth， which is unusual in theropod dinosaurs （LeBlanc et al.， 2017; Hanaiand Tsuihiji， 2018）. Additionally， tyrannosaurids may display an ontogenetic shift in toothreplacement direction in the maxilla （Carr， 2016; Hanai and Tsuihiji， 2018）. However， due tothe lack of data on non-tyrannosaurid tyrannosauroids， it remains unclear whether these toothreplacement patterns in tyrannosaurids represent adaptations to their peculiar feeding strategiesor general patterns within tyrannosauroids.

Here， we present CT data of the dentitions of two specimens of the Late Jurassictyrannosauroid Guanlong wucaii， including a juvenile and an adult （Xu et al.， 2006）.Utilizing high-resolution CT data， we reconstructed 3D models of all generations of teethand quantitatively analyzed tooth replacement patterns based on Zahnreihen reconstruction.This study provides novel insights into the tooth replacement patterns in non-tyrannosauridtyrannosauroids， offering new perspectives on the dental evolution within tyrannosauroids.

Institutional abbreviations IVPP， Institute of Vertebrate Paleontology and Paleoanthropology，Chinese Academy of Sciences， Beijing， China; IHEP， Institute of High EnergyPhysics， Chinese Academy of Sciences; MPC， Institute of Paleontology and Geology，Mongolian Academy of Sciences， Ulaanbaatar， Mongolia

2 Materials and methods

2.1 Materials

Two specimens of Guanlong wucaii were examined， including the holotype （IVPPV14531） and a referred specimen （V14532）. Both specimens were excavated from the UpperJurassic （Oxfordian） Shishugou Formation of Wucaiwan， Xinjiang， China （Xu et al.， 2022）.V14531 is an adult individual， while V14532 is a juvenile （Xu et al.， 2006）.

IVPP V14531 consists of an almost complete skull and a partially articulated postcranialskeleton. Only the upper dentition was CT scanned and 3D reconstructed， as the lower jawis not preserved. V14532 consists of a nearly complete skull with a nearly complete andarticulated postcranial skeleton. The entirety of each of the dentitions was CT scanned and 3Dreconstructed.

2.2 Computed tomography and 3D reconstruction

Scanning on both specimens was carried out using the 450 kV industry-computerizedtomography （developed by IHEP at the Key Laboratory of Vertebrate Evolution and HumanOrigins， Chinese Academy of Sciences （Wang et al.， 2019）. IVPP V14531 was scanned withbeam energy of 185 kV and a flux of 150 mA at a resolution of 45.123 μm， while V14532 wasscanned with beam energy of 150 kV and a flux of 140 mA at a resolution of 31.283 μm. Theraw CT data was completed with Mimics （Materialise Medical Co， Belgium version 25.0） for3D reconstruction and measurements， and the surface meshes were rendered via Vayu 1.0 （Lu，2023）.

2.3 Zahnreihen and Z-spacing reconstruction

Zahnreihen， a concept proposed by Edmund （1960）， was widely used to quantify thetooth replacement patterns of non-mammalian gnathostomes （Edmund， 1962， 1969; DeMarand Bolt， 1981; Fastnacht， 2008; Schwarz et al.， 2015; Sassoon et al.， 2015; Hanai and Tsuihiji，2018; Sattler and Schwarz， 2021; Hu et al.， 2022）. Each Zahnreihe consists of individual teethfrom different alveoli and generations diagonally， with the maturity of teeth decreasing distally（Hanai and Tsuihiji， 2018; He et al.， 2018）. The Zahnreihen indicates the spacing betweentwo consecutive teeth sharing the same tooth replacement stage， and “Zahnreihen-spacing”（Z-spacing） is the horizontal distance between any two adjacent teeth （Edmund， 1960; DeMar，1972， 1973; Osborn， 1972; Bolt and DeMar， 1986; Sattler and Schwarz， 2021）. A Z-spacingvalue of 2 denotes perfect alternating tooth replacement， while a Z-spacing value of 1 denotesthat the tooth replacement pattern is identical in each tooth position （DeMar and Bolt， 1981;Hanai and Tsuihiji， 2018; Brink and LeBlanc， 2023）. A Z-spacing value exceeding 2 suggestsa cephalad pattern （teeth advance from distal to mesial）， while a value below 2 indicates theopposite direction （caudad pattern， Carr， 2016）. Overall， lower Z-spacing values representrelatively faster tooth replacement and denser distribution of replacement teeth （Hanai andTsuihiji， 2018）. In this study， we adopted the methodology employed in previous research toreconstruct Zahnreihen and Z-spacing， utilizing the “replacement index” （RI） as a proxy toquantify the stages of tooth replacement （DeMar and Bolt， 1981; Fastnacht， 2008; He et al.，2018; Hu et al.， 2022）. The RI for a fully erupted functional tooth without replacement teethis designated as 1.0. When a replacement tooth emerges， its RI is determined by dividing itstotal height by the total height of its predecessor （also for second-generation replacementteeth， Supplementary File 1）. As the replacement tooth is lost， the height of the resorptionpit it caused is used to estimate its total length. Additionally， the RI of its functional toothis equivalent to that of the replacement tooth plus 1.0. In instances where the crown of thefunctional tooth is damaged or absent， the RI for the replacement tooth is calculated by dividingthe crown basal length of the replacement tooth by the maximum length of the alveolus. TheZahnreihen of each quadrant in both individuals were reconstructed into scatterplots， with thex-axis representing tooth position and the y-axis representing RI values. Z-spacing values werederived from the horizontal distance between the neighboring Zahnreihen.

3 Result

3.1 Dentition of IVPP V14531

IVPP V14531 preserves most of the left upper dentition， but several of the functionalteeth are fragmented or lost （Fig. 1）. The left premaxilla bears four alveoli， which is consistentwith most theropods （Smith， 2005）. Eight tooth positions were identified in the left maxilla，with the posterior part of the tooth row missing. Almost all functional crowns except Lpm3 inthe left premaxilla are fractured. The crowns and tips of the Lmx1 and Lmx5 show significantwear and lack smooth wear facets， likely caused by taphonomical factors rather than naturalabrasion. The condition of the root of the functional teeth in more distal tooth positions ispoor， with a fracture at the root of Lmx6， and extensive damage to the entire tooth of Lmx8.On the right side， only the most mesial three alveoli of the premaxilla remain observable，with almost all crowns of the functional teeth lost. Ex situ crown fragments were presentdistal to Rpm3， although their provenance is unclear. The premaxillary teeth of V14531， likethose of most tyrannosauroids， are transversely thickened， with the mesial carina migratingmesiolingually， exhibiting an asymmetrical salinon-shaped cross-section （Brochu， 2003;Hendrickx et al.， 2015）. The only intact crowned premaxillary tooth， Lpm3， is slightly smallerin size compared to most of the maxillary teeth. It is noted that the crown basal length of Lpm4is not significantly smaller than that of Lpm3， similar to Kileskus but unlike Tyrannosaurus（Smith， 2005; Averianov et al.， 2010）. The mesial-most maxillary tooth on the left side is slightly procumbent， while those distal to Lmx5 exhibit a retrocumbency. However， there isno significant variation in the size among the maxillary teeth. Therefore， in terms of tooth size，Guanlong exhibits a condition closer to homodonty rather than heterodonty， which is typical ofmost tyrannosaurids （Currie et al.， 1990; Brochu， 2003; Smith， 2005）.

An alternating replacement pattern of odd and even alveoli is evident in the premaxillarydentition. Odd-numbered teeth are at relatively early stages of tooth replacement， withreplacement teeth developing the tooth crown， and there is no significant resorption atthe base of the respective functional teeth. Among them， the replacement tooth of Lpm3is approximately 1/5 the total height of its functional tooth. On the right premaxilla， thereplacement teeth of Rpm1 and Rpm3 are similar in size to their counterparts on the leftside. The broken surfaces of their functional teeth did not occur at the base of the crown butrather closer to the root. Therefore， the loss of functional teeth in both cases is attributed topostmortem taphonomy rather than natural shedding. The replacement of the even-numberedteeth is at relatively later stages than the odd-numbered teeth， with the old functional teethhaving undergone significant resorption. Only one dental generation is present in Lpm2， dueto the full resorption of the previous functional tooth. The only preserved tooth in this alveolushas just erupted into the oral cavity， with an incipient root that is not yet well-defined. InLpm4， the old functional tooth was resorbed badly by the replacement tooth， with only thelabial side remaining. The replacement tooth of Lpm4 exhibits a slightly smaller tooth heightthan the new functional tooth of Lpm2， suggesting they are of contemporary generations.However， the replacement tooth of Lpm4 has not yet erupted from the dental socket， and theroot development has not commenced. Rpm2 represents the only preserved even-numberedtooth position in the right premaxilla， but it is significantly smaller in size than its counterparton the left side. The developmental positions of all the premaxillary replacement teeth aresimilar， with their current bases aligning on the same horizontal level， which is slightly moreoriented toward the root apex than the crown bases.

The maxillary teeth exhibit a similar alternating replacement pattern to the premaxillaryteeth， with the replacement stages of the even-numbered teeth relatively delayed. Among them， the replacement teeth of Lmx1 and Lmx3 are small， approximately 1/5 the total height of thefunctional teeth， while the replacement tooth of Lmx5 dramatically increases to two-fifthsthe size of the functional tooth. No replacement tooth is detected for Lmx7， only a functionaltooth with an undeveloped root. Among the even-numbered teeth， the replacement tooth ofLmx2 is well developed， approximately half the size of the functional tooth. The base of thecorresponding functional tooth is severely resorbed， and the resorption has occurred on thelabial side of the root. The functional crown is crushed towards the lingual side postmortem，partially overlapping the apex of the replacement tooth in lingual view. Further resorptionwas detected in Lmx4， with only a minor portion of the labial side of the previous functionaltooth remaining. The replacement tooth of Lmx4 represents the largest replacement tooth insize within the maxillary dentition， exceeding half the size of the adjacent functional teeth.The replacement tooth of Lmx6 is smaller， similar to that of Lmx1 and Lmx3， indicating arelatively early stage of tooth replacement. The developmental positions of the replacementteeth in the maxilla vary with tooth position. The replacement teeth of Lmx1 and Lmx6 arenotably lower than those from positions 2 to 5. Generally， the development positions of thereplacement teeth in the maxilla are significantly higher than those in the premaxilla.

3.2 Dentition of IVPP V14532

Both the upper and lower dentitions of IVPP V14532 are preserved， with the overallpreservation of the mandible better than that of the maxilla， and the preservation of the rightdentition is better than that of the left （Figs. 2， 3）. Although the right premaxillary dentitionhas only one tooth position preserved， the left premaxilla retains a complete set of four alveoli（Fig. 2）. Ten alveoli can be identified on the left maxilla， with the more posterior dentition lostdue to taphonomy. The right maxilla preserves a relatively complete tooth row， with at least15 tooth positions visible， fewer than that of Kileskus and Proceratosaurus but comparable tothat of Yutyrannus （Averianov et al.， 2010; Rauhut et al.， 2010; Xu et al.， 2012）. Both dentarieshouse 21 tooth positions， more than in Proceratosaurus. The dentary tooth row is slightlyshorter than the upper tooth row， resembling most theropods except for Proceratosaurus，whose lower dentition is significantly shorter than the upper dentition （Rauhut et al.， 2010）.The upper dentition of V14532 exhibits slight heterodonty， resembling that of V14531. Mostmesial maxillary teeth are slightly procumbent， whereas those distal to Lmx5 are retrocumbent.A similar tendency can be detected in the lower jaw. The first two dentary teeth on both sidesdisplay procumbency， with most of the mesial teeth tilting more remarkably than distal ones，in contrast to the significant procumbency of mesial dentary dentition in Proceratosaurus（Rauhut et al.， 2010）. While the dental size increases from d1 to d3， a gradual decrease occursfrom d4 to d21. Thus， d3 and d4 harbour the largest dentary teeth， resembling the condition inTyrannosaurus and Tarbosaurus （Smith， 2005; Hanai and Tsuihiji， 2018）.

The left premaxilla of V14532 has a replacement tooth or evidence of replacementteeth at every tooth position， displaying a similar odd-even alternating pattern as in V14531（Fig. 2A）. The replacement teeth of Lpm1 and Lpm3 are at relatively early stages of toothontogeny， with Lpm3 slightly larger than Lpm1 in size. The functional teeth at these positionshave not undergone significant resorption， indicating that the loss of their crowns is not due tonatural shedding. Among the even-numbered teeth， the functional teeth have endured extensiveresorption. Within Lpm2， the lingual side of the functional tooth has been resorbed， possiblyleading to the shedding of the crown. While the replacement teeth of Lpm4 were not preserved，a deep resorption pit on the lingual side of the functional tooth attests to the existence of thereplacement teeth. On the right premaxilla， only Rmx2 retains fossilized dental tissues. Itsreplacement tooth is well-developed and similar in size to the replacement tooth of Lpm2. Thelingual side of the functional tooth has undergone significant resorption， with the replacementtooth invading into the pulp cavity of the predecessor. However， the degree of resorption ofRpm2 is less than that of its counterpart on the left side. Similar to V14531， the developmentallocations of the premaxillary replacement teeth in V14532 show little variation but aregenerally lower than those of the maxillary dentition replacement teeth.

The maxillary teeth on both sides exhibit two generations of teeth in most alveoli anddisplay a complex alternating replacement pattern （Fig. 2）. In the left upper dentition， fromposition 1 to position 5， the replacement teeth in even-numbered positions are relatively largerin size， while the replacement teeth in odd-numbered positions are larger from position 6 toposition 10. Overall， the replacement teeth in all odd-numbered positions are of similar sizes（Fig. 2A）. The small-sized replacement teeth of Lmx1 and Lmx3 have caused resorptionpits on the lingual side of their functional teeth. Although the replacement tooth of Lmx5 isseverely fragmented， there is a significant resorption pit on the lingual side of its functionaltooth. It allows for an estimation of the size of this missing replacement tooth， which is likelycomparable to that of Lmx3. The resorption pit of the functional tooth of Lmx7 extends deepinto the pulp cavity， and the replacement tooth is much larger than those of Lmx1 and Lmx3.No remnants of the old functional tooth are retained in Lmx9， possibly indicating completeresorption. Among the even-numbered teeth， both Lmx2 and Lmx4 show late stages of toothreplacement， with their replacement teeth well developed. While the functional crown ofLmx2 has not shed， the old functional tooth of Lmx4 is almost fully resorbed. Although thereplacement tooth of Lmx4 has erupted into the oral cavity， it has not yet begun to developthe root， similar to the condition of Lmx4 in V14531. The even-numbered alveoli from Lmx6to Lmx10 share a similar stage of tooth replacement， although both their functional andreplacement teeth are poorly preserved. By estimating the size of the resorption pit， it can beinferred that the replacement tooth of Lmx8 is similar in size to those of Lmx6 and Lmx10，indicating an early stage of replacement tooth development. There is a noticeable transitionin the stage of tooth replacement between Lmx4 and Lmx6， possibly indicating a shift fromthe end of one tooth replacement cycle to the beginning of a new one. The basal length of thefunctional teeth of Lmx6 is similar to that of Lmx4， suggesting that they are contemporary.Therefore， in the even-numbered teeth， the stages of tooth replacement gradually become moremature from proximal to distal.

The right upper jaw preserves a more complete tooth row and provides more informationon tooth replacement （Fig. 2B）. The replacement pattern of Rmx1 to Rmx10 is generallysimilar to that of the left upper jaw， with replacement teeth in even-numbered positions closeto Rmx6 being more developed， while those close to Rmx6 in odd-numbered positions haveless mature replacement teeth. Positions 12 to 15 again show a pattern of relatively moremature replacement teeth in the even-numbered positions. No replacement teeth are observedin Rmx1 and Rmx5， with the functional tooth of Rmx5 showing a resorption pit on the lingualside of the root. However， in Rmx1， there were no resorption pits on the functional tooth orthe crypt on the lingual side of the alveolus detected， indicating that the replacement toothhas not yet been initiated in this tooth position. Based on the size of this resorption pit， it canbe speculated that the replacement tooth of Rmx5 is developing， with a size between that ofRmx3 and Rmx7. Thus， the replacement teeth at positions 1， 3， and 5 are in relatively earlystages of tooth replacement， with resorption having not yet begun to invade deeply into thepulp cavity. Rmx2 and Rmx4， on the other hand， exhibit significant resorption and haveformed resorption trenches on the lingual side of their functional teeth. The replacement toothof Rmx2 is approximately 1/4 the size of the functional tooth， but resorption has occurredon the labial side of its precursor. The replacement tooth of Rmx4 has reached about 2/5the size of the functional tooth and has completely resorbed the root apex of the functionaltooth. However， the degree of resorption in both Rmx2 and Rmx4 is not as pronounced asin their counterparts on the left side. Rmx6 has only one tooth that has just erupted into theoral cavity， with its root not clearly defined， revealing a newly attached functional tooth. Thereplacement teeth at positions 8， 10， and 12 are similar in size and stage of tooth replacement.While the replacement tooth of Rmx14 shares a similar size to the replacement teeth in thesethree alveoli， the replacement stage of Rmx14 is significantly more mature than that of thoseat positions 8， 10， and 12. The replacement tooth of Rmx14， which is small in size， has fullyinvaded the pulp cavity of the functional tooth， leading to the shedding of the functional toothcrown. The replacement teeth of Rmx7 and Rmx9 are well developed， with their functionalteeth severely resorbed. While the functional tooth of Rmx7 has not yet shed， remnants of thefunctional tooth remain in Rmx9. The replacement teeth of Rmx7 and Rmx9 are similar insize to the functional tooth crowns of Rmx13 and Rmx15， suggesting that they are at distinctstages of the same tooth replacement generation. The functional teeth of Rmx13 and Rmx15are highly mature， with Rmx13 showing a replacement tooth and a resorption pit， while thepulp cavity of Rmx15 is completely closed， indicating a late stage of tooth ontogeny. Thedevelopmental position of the upper jaw teeth replacement of V14532 is slightly different fromthat of V14531. The bases of the replacement teeth mesial to position 7 on both sides are atsimilar levels， while those distal to position 7 are deeper into the jawbone.

The dentary dentition of V14532 is characterized by the presence of three toothgenerations within one alveolus （Fig. 3）. On the right side， a second-generation replacementtooth was detected in Rd4， with the old functional tooth still active （Figs. 3， 4）. The leftdentary， however， houses three even-numbered alveoli （Ld2， Ld4 and Ld6） developing secondgenerationreplacement teeth （Fig. 3A）. Meanwhile， the crowns of the old functional teeth ofthese three alveoli have newly shed but the roots remain， with the first-generation replacementteeth having erupted into the oral cavity. Within the right dentary， the odd-numbered teethgenerally exhibit relatively similar stages of tooth replacement， even though no replacementteeth are observed at Rd1 due to preservation （Fig. 3B）. The stages of tooth replacementfrom Rd3 to Rd11 are relatively similar. Their replacement teeth are at relatively early stages，causing resorption pits on the lingual side of the functional teeth， but they are not yet invadinginto the pulp cavity. Rd13 to Rd21 demonstrate a complete process of replacement teeth thatgradually increasing in size and becoming more mature until they become functional teethand form a new generation of replacement teeth. The replacement tooth of Rd13 is larger thanthat of Rd11 in size and invades into the pulp cavity of the functional tooth. The replacementtooth of Rd15 has erupted into the oral cavity but has not yet fully resorbed the remnants ofthe previous tooth generation. The new functional teeth of Rd17 and Rd19 have developedclear roots without replacement teeth. A new tooth generation was established in Rd21， withthe root of the functional tooth fully developed. On the other hand， the replacement patternof the even-numbered teeth is similar to that of the odd-numbered teeth. Apart from Rd4，which develops a second-generation replacement tooth， the functional teeth of Rd2 to Rd8exhibit stages of replacement similar to those of first-generation replacement teeth， showingsignificant resorption on the lingual side of the functional teeth. In Rd10， the only preservedtooth belongs to the same generation as the first-generation replacement teeth of Rd2 to Rd8，with the previous generation of functional teeth having been completely resorbed. The maturityof the replacement teeth from Rd10 to Rd20 gradually increases distally. The functional toothof Rd18 has undergone significant resorption， forming a resorption pit. No remnants of theprevious generation of the functional tooth are observed in Rd20， possibly indicating completeresorption. The new functional teeth have erupted into the oral cavity but have not yet begun toform distinct roots.

The dental arch of the left side exhibits a similar replacement pattern to the right side， butall the left side positions show relatively more mature stages of tooth replacement comparedto their counterparts on the right side. For example， the replacement teeth on the left side aregenerally larger than those of the right side （e.g. the odd-numbered positions from 3 to 11），and while the old teeth on the right side are still functional， their counterparts on the left havealready shed due to resorption （e.g. position 2， 4 and 13）. The developmental positions ofthe replacement teeth in the lower jaw show a trend of initial decreasing and then increasingtowards the distal end.

3.3 Zahnreihen and Z-spacing reconstruction

To quantify the tooth replacement patterns of Guanlong， we reconstruct the Zahnreihenand Z-spacing of the left upper jaw of IVPP V14531 and all jaw quadrants of V14532. Theresults show that the upper dentitions of both individuals and the lower dentition of V14532 allexhibit mean Z-spacing values exceeding 2 （Figs. 5–7）. If the Zahnreihen of the premaxillaryand the maxillary teeth are calculated separately， the mean Z-spacing value of the left maxillarydentition （2.50） is dramatically larger than that of the left premaxillary dentition （2.07） in the juvenile （V14532， Fig. 6A）. This difference in Z-spacing values between the premaxillaryand the maxillary teeth was also detected in V14531， with 1.82 for premaxillary dentition and2.69 for maxillary dentition （Fig. 5）. Furthermore， there is a decrease of the Z-spacing value inthe left premaxillary dentition （2.07 versus 1.82）， and an increase in the left maxilla dentition（2.50 versus 2.69） throughout ontogeny （Figs. 5， 6A）. The right maxillary dentition of V14532shares a similar Z-spacing value （2.54， Fig. 6B） with its left counterpart （2.50， Fig. 6A）. Whilethe Z-spacing values for both dentaries in V14532 are similar， the value of the left side （2.12，F(xiàn)ig. 7A） is slightly lower than that of the right side （2.23， Fig. 7B）， which is possibly due tothe relatively lower Z-spacing value resulting from more second-generation replacement teethon the left side.

4 Discussion

Like most non-mammalian vertebrates， Guanlong exhibits an iguanian or subdentalreplacement pattern with replacement teeth developing on the lingual side of the functionaltooth within the individual alveolus （Edmund， 1960）. In contrast， extant crocodilians andtoothed birds develop their replacement teeth entirely beneath their functional teeth， with thepulp cavity of the functional tooth being invaded at an early stage of tooth replacement （Dumontet al.， 2016; Fong et al.， 2016; LeBlanc et al.， 2017; Hanai and Tsuihiji， 2018; Kundrát et al.，2019; Wu et al.， 2021）. While Guanlong does exhibit some replacement teeth completelyunderneath the functional teeth （e.g. Rd2 and Rd8 of V14532） like birds and crocodiles， theseteeth constitute only a small fraction of the total tooth count （approximately 5% in the lowerright dentition of V14532， Fig. 3B）. Lawson et al. （1971） hypothesized that the number of teethin each stage reflects the relative duration of that stage in the replacement cycle. Consequently，the majority of the replacement teeth in Guanlong spend most of their formation time outsidethe pulp cavity of their predecessors， resembling patterns seen in other non-avian theropodsrather than those seen in birds and crocodilians. Remarkably， the subsequent stage， in whichthe functional tooth sheds leaving only the replacement tooth in the alveolus， also representsa negligible proportion of the tooth count （about 8% in the lower right dentition of V14532，F(xiàn)ig. 3B）. This suggests that the rapid processes of tooth shedding and attachment observed inGuanlong align with the broader pattern documented in Tarbosaurus， crocodilians， and evenamphibians （Lawson et al.， 1971; Hanai and Tsuihiji， 2018）.

Guanlong represents one of the few theropods developing second-generation replacementteeth. So far， second-generation replacement teeth have only been reported in Tarbosaurus，Gorgosaurus， Oxalaia， Allosaurus， and Majungasaurus （Edmund， 1960; Kellner et al.，2011; LeBlanc et al.， 2017; Hanai and Tsuihiji， 2018; D’Emic et al.， 2019）. However， thedistributional pattern of second-generation replacement teeth in respective individualsremains to be only vaguely understood. Second-generation replacement teeth were found inthe premaxilla of Oxalaia and in the maxillae of Majungasaurus and Allosaurus. Due to theincomplete dental remains of the previously examined fossils， it cannot be determined if thesegroups have a more widespread distribution of second-generation replacement teeth in otherjaw elements. MPC-D 107/7， a juvenile Tarbosaurus specimen， represents the most completematerial with records of second-generation replacement teeth to date （Hanai and Tsuihiji，2018）. Second-generation replacement teeth are present in both the maxillary and the dentary，primarily in the middle region of the tooth rows. In Guanlong， second-generation replacementteeth are only present in the lower jaw of the juvenile （V14532）. It is unclear whether or notthe adult Guanlong has second-generation replacement teeth due to the absence of the lowerjaw in V14531 （Fig. 1）. In the lower jaw of V14532， the second-generation replacement teethare limited to the mesial alveoli， contrasting with the widespread distribution in Tarbosaurus.On the right dentary， only Rd4， which is associated with the largest functional tooth in thelower jaw， develops the second-generation replacement tooth （Fig. 3B）. It suggests that d4 mayinitiate its replacement tooth earlier than other dentary alveoli.

This is also evident in the left dentary， where Ld4 exhibits the largest second-generationreplacement tooth. Both Ld2 and Ld6 also develop the same generation of replacement teeth，but the old functional tooth of Ld2 has been fully resorbed， while only a small amount ofresidual dental tissue from the previous functional tooth remains in Ld6. Ld4， on the otherhand， retains relatively more dental remnants from the old functional tooth， suggesting arelatively late shedding and a potential prolonged dental life-span. Therefore， at least inGuanlong and Tarbosaurus， second-generation replacement teeth are more likely to beassociated with the largest functional teeth. This may be because larger functional teeth takemore time to be completely resorbed， thus having longer lifespans， which makes it more likelyfor three generations of teeth to coexist. The potential association between second-generationreplacement teeth and the largest functional teeth may also represent a functional adaptation，whereby the largest functional teeth also take the longest time to develop to full functionality.The development of second-generation replacement teeth promotes the reduction of thefunctional gaps that are created by shedding or the accidental loss of large functional teeth.

The direction of tooth replacement and Z-spacing are pivotal features for characterizingtooth replacement patterns， albeit there are ongoing debates regarding the mechanisms behind Zahnreihen formation and maintenance （Edmund， 1960， 1962， 1969; Osborn， 1972， 1974，1977; Westergaard and Ferguson， 1986， 1987， 1990）. Typically， tooth replacement in mostnon-mammalian jawed vertebrates follows a cephalad pattern， accompanied by Z-spacingvalues exceeding 2 （Edmund， 1960）. Interestingly， some species exhibit an ontogenetic shift intooth replacement direction， correlated with variations in Z-spacing. For instance， crocodilesshift their tooth replacement direction from cephalad to caudad with a decrease in Z-spacingthroughout ontogeny， which is potentially linked to jaw growth occurring without altering thetooth count （Edmund， 1962; Carr， 2016; Brink and LeBlanc， 2023）. Tyrannosaurids representa rare fossil group that is possibly undergoing an ontogenetic shift of tooth replacementdirection from caudad to cephalad in maxillary dentition with an increase of Z-spacing values，despite the mechanisms behind it remaining unclear （Carr， 2016; Hanai and Tsuihiji， 2018）.According to Zahnreihen reconstructions， both specimens of Guanlong exhibit consistentcephalad patterns in the maxillary， and the dentary dentitions have Z-spacing values surpassing2.0. However， an increase in Z-spacing value was noted in the left maxillary dentition asGuanlong matured， resembling an ontogenetic variation similar to that of tyrannosaurids. Thissuggests that the ontogenetic increase in Z-spacing may represent a plesiomorphic trait forTyrannosauroidea or a broader clade.

Different tooth replacement patterns have been observed between the premaxilla andthe maxilla in the Tyrannosauridae， which is possibly associated with their heterodonty andfunction differentiation （Reichel， 2010; Hanai and Tsuihiji， 2018）. While the small incisiformpremaxillary teeth are likely used for muscle tissue extraction， the robust maxillary teeth arelinked to deep bites （Hone and Watabe， 2010）. Occasionally， a functional gap arises whentwo successive alveoli shed at the premaxilla-maxilla boundary， due to the discontinuity intooth replacement waves （Hanai and Tsuihiji， 2018）. However， in the examined specimensof Guanlong， premaxillary teeth share a similar alternating replacement pattern to the mesialmaxillary teeth， where the even-numbered teeth are relatively more mature. It reveals that inGuanlong， at least one functional tooth exists in either the premaxilla or the maxilla at thepremaxilla-maxilla boundary， thus preventing the functional gap. Nonetheless， an unusualdecrease in the Z-spacing values between the premaxillary and maxillary Zahnreihen inthe upper left dentitions of both specimens indicates the discontinuity of the replacementwaves at the premaxilla-maxilla boundary in both specimens （Figs. 5， 6A）. Additionally， thepremaxillary teeth of Guanlong display notably lower Z-spacing values than the maxillaryteeth， further supporting the presence of this discontinuity. Throughout ontogeny， thisdiscontinuity may amplify the difference in the Z-spacing values between the premaxillaryand maxillary teeth， with 0.43 （2.07 versus 2.50， Fig. 6A） and 0.87 （1.82 versus 2.69， Fig. 5）for V14532 and V14531， respectively. While the exact mechanism behind this discontinuityremains uncertain， the evidence supports the conclusion that the premaxillary and the maxillarydentitions were replaced by separate processes， predating the origin of the Tyrannosauridae（Hanai and Tsuihiji， 2018）.

Acknowledgements This study was supported by the National Natural Science Foundationof China （42288201， 42372031， 41972025） and the International Partnership Program ofChinese Academy of Sciences （Grant No. 132311KYSB20190010）. The funders had no rolein study design， data collection and analysis， the decision to publish， or the preparation of themanuscript. We declare that AI-assisted technologies were utilized in creating this article.

摘要：暴龍科是研究最為深入的獸腳類支系之一，發(fā)育極端增厚的齒列，可能具有獨(dú)特的碎骨捕食策略。與進(jìn)食行為聯(lián)系緊密的牙齒替換模式在這一類群中得到了廣泛的研究，但前人對(duì)早期分異的暴龍超科成員卻鮮有相應(yīng)的研究。利用高分辨率CT數(shù)據(jù)，三維重建了晚侏羅世暴龍超科——五彩冠龍（Guanlong wucaii）兩件標(biāo)本的齒列，為研究暴龍超科的牙齒替換模式提供了新的信息。五彩冠龍幼年個(gè)體的下頜齒列發(fā)育二代替換齒，這一特征此前僅在包括暴龍科在內(nèi)的大型獵食性獸腳類恐龍中有報(bào)道。Zahnreihen重建顯示，五彩冠龍的兩個(gè)個(gè)體在上頜骨齒和齒骨齒中均表現(xiàn)出Z間距大于2.0的從前向后的波狀替換模式。五彩冠龍上頜骨齒的Z間距隨著個(gè)體發(fā)育而變大，與在暴龍科成員中觀察到的個(gè)體發(fā)育變化類似。此外，五彩冠龍?jiān)谇吧项M骨-上頜骨交界處顯示出了與特暴龍（Tarbosaurus）相似的牙齒替換波的間斷，且這種替換波的間斷會(huì)隨著個(gè)體發(fā)育加劇。這表明，一些與暴龍科類似的牙齒替換模式在暴龍科起源之前就已經(jīng)產(chǎn)生。

關(guān)鍵詞：晚侏羅世，暴龍超科，牙齒替換，三維重建

中圖法分類號(hào)：Q915.864 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：2096–9899（2024）03–0225–20

References

Abler W L， 1992. The serrated teeth of tyrannosaurid dinosaurs， and biting structures in other animals. Paleobiology， 18：161–183

Averianov A O， Krasnolutskii S A， Ivantsov S V， 2010. A new basal coelurosaur （Dinosauria： Theropoda） from the Middle Jurassic of Siberia. Proc Zool Inst Russ Acad Sci， 314： 42–57

Bolt J R， DeMar R E， 1986. Computer simulation of tooth replacement with growth in lower tetrapods. J Vert Paleont， 6：233–250

Brink K S， LeBlanc A R， 2023. How the study of crocodylian teeth influences our understanding of dental development，replacement， and evolution in dinosaurs. In： Araújo R， Fernandez V， Johnston P S eds. Rudling Reptiles： Crocodylian Biology and Archosaur Paleobiology. Bloomington： Indiana University Press. 240–257

Brochu C A， 2003. Osteology of Tyrannosaurus rex： insights from a nearly complete skeleton and high-resolution computed tomographic analysis of the skull. J Vert Paleont， 22： 1–138

Brusatte S L， Carr T D， 2016. The phylogeny and evolutionary history of tyrannosauroid dinosaurs. Sci Rep， 6： 20252

Brusatte S L， Norell M A， Carr T D et al.， 2010. Tyrannosaur paleobiology： new research on ancient exemplar organisms.Science， 329： 1481–1485

Buckley L G， Larson D W， Reichel M et al.， 2010. Quantifying tooth variation within a single population of Albertosaurus sarcophagus （Theropoda： Tyrannosauridae） and implications for identifying isolated teeth of tyrannosaurids. Can J Earth Sci， 47： 1227–1251

Carr T D， 2016. Toward understanding phylogenetic and ontogenetic tooth loss in Archosauria： documenting patterns of tooth replacement in Tyrannosaurus rex. Paper presented at the 76th Annual Meeting of the Society of Vertebrate Paleontology. Utah： Grand America Hotel. 110

Carr T D， 2020. A high-resolution growth series of Tyrannosaurus rex obtained from multiple lines of evidence. PeerJ， 8：e9192

Currie P J， 2003. Cranial anatomy of tyrannosaurid dinosaurs from the Late Cretaceous of Alberta， Canada. Acta Palaeontol Pol， 48： 191–226

Currie P J， Rigby J K， Sloan R E， 1990. Theropod teeth from the Judith River Formation of southern Alberta， Canada. In：Carpenter K， Currie P J eds. Dinosaur Systematics： Approaches and Perspectives. Cambridge： Cambridge University Press. 107–125

Dalman S G， Loewen M A， Pyron R A et al.， 2023. A giant tyrannosaur from the Campanian–Maastrichtian of southern North America and the evolution of tyrannosaurid gigantism. Sci Rep， 13： 22124

DeMar R， 1972. Evolutionary implications of Zahnreihen. Evolution， 26： 435–450

DeMar R， 1973. The functional implications of the geometrical organization of dentitions. J Vert Paleont， 47： 452–461

DeMar R， Bolt J R， 1981. Dentitional organization and function in a Triassic reptile. J Vert Paleont， 55： 967–984

DePalma II R A， Burnham D A， Martin L D et al.， 2013. Physical evidence of predatory behavior in Tyrannosaurus rex. Proc Natl Acad Sci， 110： 12560–12564

Dumont M， Tafforeau P， Bertin T et al.， 2016. Synchrotron imaging of dentition provides insights into the biology of Hesperornis and Ichthyornis， the “l(fā)ast” toothed birds. BMC Evol Biol， 16： 178

D’Emic M D， O’Connor P M， Pascucci T R et al.， 2019. Evolution of high tooth replacement rates in theropod dinosaurs.PLoS One， 14： e0226897

Edmund A G， 1960. Tooth replacement phenomena in the lower vertebrates. R Ont Mus， 52： 1–190

Edmund A G， 1962. Sequence and rate of tooth replacement in the Crocodilia. R Ont Mus， 56： 1–42

Edmund A G， 1969. Dentition. In： Gans C， Bellairs A A， Parson T S eds. Biology of the Reptilia. London and New York：Academic Press. 117–200

Erickson G M， 1995. Split carinae on tyrannosaurid teeth and implications of their development. J Vert Paleont， 15： 268–Erickson G M， 1996. Incremental lines of von Ebner in dinosaurs and the assessment of tooth replacement rates using growth line counts. Proc Natl Acad Sci， 93： 14623–14627

Erickson G M， Olson K H， 1996. Bite marks attributable to Tyrannosaurus rex： preliminary description and implications. J Vert Paleont， 16： 175–178

Erickson G M， Kirk S D V， Su J et al.， 1996. Bite-force estimation for Tyrannosaurus rex from tooth-marked bones. Nature，382： 706–708

Farlow J O， Brinkman D L， 1994. Wear surfaces on the teeth of tyrannosaurs. The Paleontological Society Special Publications. Cambridge： Cambridge University Press. 165–176

Fastnacht M， 2008. Tooth replacement pattern of Coloborhynchus robustus （Pterosauria） from the Lower Cretaceous of Brazil. J Morphol， 269： 332–348

Fong R K， LeBlanc A R， Berman D S et al.， 2016. Dental histology of Coelophysis bauri and the evolution of tooth attachment tissues in early dinosaurs. J Morphol， 277： 916–924

Funston G F， Powers M J， Whitebone S A et al.， 2021. Baby tyrannosaurid bones and teeth from the Late Cretaceous of western North America. Can J Earth Sci， 58： 756–777

Gignac P M， Erickson G M， 2017. The biomechanics behind extreme osteophagy in Tyrannosaurus rex. Sci Rep， 7：2012

Hanai T， Tsuihiji T， 2018. Description of tooth ontogeny and replacement patterns in a juvenile Tarbosaurus bataar （Dinosauria： Theropoda） using CT-scan data. Anat Rec， 302： 1210–1225

He Y， Makovicky P J， Xu X et al.， 2018. High-resolution computed tomographic analysis of tooth replacement pattern of the basal neoceratopsian Liaoceratops yanzigouensis informs ceratopsian dental evolution. Sci Rep， 8： 5870

Hendrickx C， Mateus O， Araújo R， 2015. A proposed terminology of theropod teeth （Dinosauria， Saurischia）. J Vert Paleont，35： e982797

Holtz Jr T R， 2021. Theropod guild structure and the tyrannosaurid niche assimilation hypothesis： implications for predatory dinosaur macroecology and ontogeny in later Late Cretaceous Asiamerical. Can J Earth Sci， 58： 778–795

Hone D W， Watabe M， 2010. New information on scavenging and selective feeding behaviour of tyrannosaurids. Acta Palaeontol Pol， 55： 627–634

Horner J R， Padian K， 2004. Age and growth dynamics of Tyrannosaurus rex. Proc R Soc B， 271： 1875–1880

Hu J， Forster C A， Xu X et al.， 2022. Computed tomographic analysis of the dental system of three Jurassic ceratopsians and implications for the evolution of tooth replacement pattern and diet in early-diverging ceratopsians. Elife， 11：e76676

Hurum J R H， Currie P J， 2000. The crushing bite of tyrannosaurids. J Vert Paleont， 20： 619–621

Hurum J R H， Sabath K， 2003. Giant theropod dinosaurs from Asia and North America： skulls of Tarbosaurus bataar and Tyrannosaurus rex compared. Acta Palaeontol Pol， 48： 161–190

Kellner A W， Azevedo S A， Machado E B et al.， 2011. A new dinosaur （Theropoda， Spinosauridae） from the Cretaceous （Cenomanian） Alcantara Formation， Cajual Island， Brazil. An Acad Bras Cienc， 83： 99–108

Kundrát M， Nudds J， Kear B P et al.， 2019. The first specimen of Archaeopteryx from the Upper Jurassic M?rnsheim Formation of Germany. Hist Biol， 31： 3–63

Lambe L M， 1917. The Cretaceous theropodus dinosaur Gorgosaurus. Mem Geol Surv Canada， 100： 1–84

Lautenschlager S， Witmer L M， Altangerel P et al.， 2014. Cranial anatomy of Erlikosaurus andrewsi （Dinosauria，Therizinosauria）： new insights based on digital reconstruction. J Vert Paleont， 34： 1263–1291

Lawson R， Wake D B， Beck N T， 1971. Tooth replacement in the red‐backed salamander， Plethodon cinereus. J Morphol，134： 259–269

LeBlanc A R， Brink K S， Cullen T M et al.， 2017. Evolutionary implications of tooth attachment versus tooth implantation： a case study using dinosaur， crocodilian， and mammal teeth. J Vert Paleont， 37： e1354006

Lu J， 2023. Vayu 1.0， a new set of tools for visualizing surface meshes. Vert PalAsiat， 61： 71–80

Maho T， Reisz R R， 2024. Exceptionally rapid tooth development and ontogenetic changes in the feeding apparatus of the Komodo dragon. PLoS One， 19： e0295002

Maho T， Maho S， Scott D et al.， 2022. Permian hypercarnivore suggests dental complexity among early amniotes. Nat Commun， 13： 4882

Marshall C R， Latorre D V， Wilson C J et al.， 2021. Absolute abundance and preservation rate of Tyrannosaurus rex. Science，372： 284–287

Meers M B， 2002. Maximum bite force and prey size of Tyrannosaurus rex and their relationships to the inference of feeding behavior. Hist Biol， 16： 1–12

Osborn H F， 1905. Article XIV.-Tyrannosaurus and other Cretaceous carnivorous dinosaurs. Bull Am Mus Nat Hist， 21：259–265

Osborn H F， Brown B， 1906. Tyrannosaurus， Upper Cretaceous carnivorous dinosaur： second communication. Bull Am Mus Nat Hist， 22： 281–296

Osborn J W， 1972. On the biological improbability of Zahnreihen as embryological units. Evolution， 26： 601–607

Osborn J W， 1974. On the control of tooth replacement in reptiles and its relationship to growth. J Theor Biol， 46： 509–527

Osborn J W， 1977. The interpretation of patterns in dentitions. Zool J Linn Soc， 9： 217–229

Owocki K， Kremer B， Cotte M et al.， 2020. Diet preferences and climate inferred from oxygen and carbon isotopes of tooth enamel of Tarbosaurus bataar （Nemegt Formation， Upper Cretaceous， Mongolia）. Palaeogeogr Palaeoclimatol Palaeoecol， 537： 109190

Paul G S， Persons W S， Van Raalte J， 2022. The tyrant lizard king， queen and emperor： multiple lines of morphological and stratigraphic evidence support subtle evolution and probable speciation within the North American genus Tyrannosaurus. Evol Biol， 49： 156–179

Peterson J E， Tseng Z J， Brink S， 2021. Bite force estimates in juvenile Tyrannosaurus rex based on simulated puncture marks. PeerJ， 9： e11450

Powers M J， Fabbri M， Doschak M R et al.， 2021. A new hypothesis of eudromaeosaurian evolution： CT scans assist in testing and constructing morphological characters. J Vert Paleont， 41： e2010087

Rauhut O W， Milner A C， Moore-Fay S， 2010. Cranial osteology and phylogenetic position of the theropod dinosaur Proceratosaurus bradleyi （Woodward， 1910） from the Middle Jurassic of England. Zool J Linn Soc， 158： 155–195

Rayfield E J， 2004. Cranial mechanics and feeding in Tyrannosaurus rex. Proc R Soc B， 271： 1451–1459

Reichel M， 2010. The heterodonty of Albertosaurus sarcophagus and Tyrannosaurus rex： biomechanical implications inferred through 3-D models. Can J Earth Sci， 49： 1253–1261

Reichel M， 2012. The variation of angles between anterior and posterior carinae of tyrannosaurid teeth. Can J Earth Sci， 49：477–491

Rowe A J， Snively E， 2022. Biomechanics of juvenile tyrannosaurid mandibles and their implications for bite force：evolutionary biology. Anat Rec， 305： 373–392

Sassoon J， Foffa D， Marek R， 2015. Dental ontogeny and replacement in Pliosauridae. R Soc Open Sci， 2： 150384

Sattler F， Schwarz D， 2021. Tooth replacement in a specimen of Tyrannosaurus rex （Dinosauria， Theropoda） from the Hell Creek Formation （Maastrichtian）， Montana. Hist Biol， 33： 949–972

Scherer C R， Voiculescu-Holvad C， 2024. Reanalysis of a dataset refutes claims of anagenesis within Tyrannosaurus-line tyrannosaurines （Theropoda， Tyrannosauridae）. Cretaceous Res， 155： 105780

Schwarz D， Kosch J C， Fritsch G et al.， 2015. Dentition and tooth replacement of Dicraeosaurus hansemanni （Dinosauria，Sauropoda， Diplodocoidea） from the Tendaguru Formation of Tanzania. J Vert Paleont， 35： e1008134

Smith J B， 2005. Heterodonty in Tyrannosaurus rex： implications for the taxonomic and systematic utility of theropod dentitions. J Vert Paleont， 25： 865–887

Therrien F， Zelenitsky D K， Voris J T et al.， 2021. Mandibular force profiles and tooth morphology in growth series of Albertosaurus sarcophagus and Gorgosaurus libratus （Tyrannosauridae： Albertosaurinae） provide evidence for an ontogenetic dietary shift in tyrannosaurids. Can J Earth Sci， 58： 812–828

Therrien F， Zelenitsky D K， Tanaka K et al.， 2023. Exceptionally preserved stomach contents of a young tyrannosaurid reveal an ontogenetic dietary shift in an iconic extinct predator. Sci Adv， 9： eadi0505

Tsuihiji T， Watabe M， Tsogtbaatar K et al.， 2011. Cranial osteology of a juvenile specimen of Tarbosaurus bataar （Theropoda， Tyrannosauridae） from the Nemegt Formation （Upper Cretaceous） of Bugin Tsav， Mongolia. J Vert Paleont， 31：497–517

Wang Y F， Wei C F， Que J M et al.， 2019. Development and applications of paleontological computed tomography. Vert PalAsiat， 57： 84–92

Westergaard B， Ferguson M W J， 1986. Development of the dentition in Alligator mississippiensis： early embryonic development in the lower jaw. J Zool， 210： 575–597

Westergaard B， Ferguson M W J， 1987. Development of the dentition in Alligator mississippiensis： later development in the lower jaws of embryos， hatchlings and young juveniles. J Zool， 212： 191–222

Westergaard B， Ferguson M W J， 1990. Development of the dentition in Alligator mississippiensis： upper jaw dental and craniofacial development in embryos， hatchlings， and young juveniles， with a comparison to lower jaw development.Am J Anat， 187： 393–421

Winkler D E， Iijima M， Blob R W et al.， 2022. Controlled feeding experiments with juvenile alligators reveal microscopic dental wear texture patterns associated with hard-object feeding. Front Ecol Evol， 10： 957725

Woodward H N， Tremaine K， Williams S A et al.， 2020. Growing up Tyrannosaurus rex： osteohistology refutes the pygmy “Nanotyrannus” and supports ontogenetic niche partitioning in juvenile Tyrannosaurus. Sci Adv， 6： eaax6250

Wu Y H， Chiappe L M， Bottjer D J et al.， 2021. Dental replacement in Mesozoic birds： evidence from newly discovered Brazilian enantiornithines. Sci Rep， 11： 19349

Xu X， Clark J M， Forster C A et al.， 2006. A basal tyrannosauroid dinosaur from the Late Jurassic of China. Nature， 439：715–718

Xu X， Wang K， Zhang K et al.， 2012. A gigantic feathered dinosaur from the Lower Cretaceous of China. Nature， 484：92–95

Xu X， Clark J M， Eberth D A et al.， 2022. The Shishugou Fauna of the Middle‐Late Jurassic transition period in the Junggar Basin of western China. Acta Geol Sin， 96： 1115–1135

國(guó)家自然科學(xué)基金（批準(zhǔn)號(hào)：42288201， 42372031， 41972025）和中國(guó)科學(xué)院國(guó)際伙伴計(jì)劃項(xiàng)目（編號(hào)：132311KYSB20190010）資助。