Supporting Information for The modern pattern of insect herbivory predates the advent of angiosperms by 60 million years Lifang Xiao, Liang Chen, Conrad C. Labandeira, Lauren Azevedo-Schmidt, Yongjie Wang, Dong Ren Conrad C. Labandeira, Yongjie Wang, Dong Ren Email: labandec@si.edu, wangyjosmy@gmail.com, rendong@cnu.edu.cn This PDF file includes: Supporting Information Text Figures S1 to S4 Tables S1 to S3 Appendices S1 to S12 Supporting Information References Supporting Information Text Overview. In these supplemental data we provide four figures, three tables, eleven appendices, and their cited references that extend the content provided in the main text. Figures S1 and S2, are photo montages that illustrate typical damage from the mid-Mesozoic plant assemblages at Daohugou, Inner Mongolia, northeastern China (late Middle Jurassic, 165 Ma); Dawangzhangzi, Liaoning, northeastern China (earlier Early Cretaceous, 125 Ma), and Rose Creek, Nebraska, U.S.A. (late Early Cretaceous, 103 Ma). A beta analysis, using a different approach than that of the main text, is provided in Figure S3. (A second rerunning of the beta analysis emphasizing a pooled dataset approach can be found as Figure 2 of the main text.). Figure S4 illustrates the most herbivorized plant hosts from five plant assemblages ranging from the Late Pennsylvanian to the late Early Cretaceous. Table S1 is an extensive dataset that lists important data for plant assemblages that is the source of our beta analysis. This matrix has formatted data displaying locality, age, functional feeding group (FFG), damage type (DT) occurrences, herbivorized surface area, and other quantitative data of the raw 180 and culled 134 plant assemblages. These data also can be found separately in the supporting materials at Mendeley Dataset S1. Table S2 provides average values of turnover and nestedness for plant assemblages from geological time intervals, including the three important mid-Mesozoic plant assemblages, related to Figure S3 above. Table S3 provides the extended definitions and descriptions of turnover and nestedness for a beta diversity analysis. Appendix S1 presents several hypotheses that explain the ecological and evolutionary rise of angiosperms during the Early Cretaceous. Appendix S2 provides a brief history of arthropod and pathogen interactions with plants since the Early Devonian, documenting major patterns during the past 420 million years. Appendix S3 gives a traditional account of the mid Cretaceous origin of modern herbivory with the ecological expansion of angiosperms and is compared to the much earlier Jurassic diversification of gymnosperms. An extended explanation of damage type and functional feeding group patterns documented in the main text is given in Appendix S4. The results of the initial running of the beta analysis emphasizing a time-series approach is provided in Figure S3 (above), explained in Appendix S5 and in Table S2. Appendix S6 extends the implications of the NMDS study. In Appendix S7, a discussion ensues of the valid uses of the leaf mass per area metric for only certain plant groups that is not available for gymnosperms. In Appendix S8, an account about trends in damage type functional breadth, explaining the “generalization” to “specialization” continuum for a DT’s presence on one or more plant hosts. The five major characteristics of Jurassic gymnosperm plant assemblages similar to those of extant angiosperm assemblages is discussed in Appendix S9. These similarities are growth form diversity, pollination mutualisms, antiherbivore plant defenses, plant– insect mimicry, and measures of plant–insect “specialization”. Appendix S10 details post 2007 updates for DTs to the Damage Guide. Appendix S11 provides the rationale for feeding event occurrence data that establishes links between damage types and their plant hosts. Last, Appendix S12 lists the data collection, standards, and five criteria necessary for acceptance of a plant assemblage into our analyses and provides the 12 data items for which (ideally) every plant assemblage should contain. Figure S1. Representative insect richness from the mid-Mesozoic plant assemblages at Daohugou northeastern China Daohugou damage. Insect damage richness from the Middle Jurassic, gymnosperm-dominated plant assemblage at Daohugou China. At (A) is margin feeding with cuspules of DT13 on Anomozamites villosus (CNU-BEN-DHG-2011676c/p), of which details of herbivory scars are enlarged at (B), together with dense trichomes indicated by arrows. Hole feeding DT4 on Anomozamites villosus (CNU-BEN-DHG-2011706) is at (C), with thick cuticle also preserved on the same leaf specimen. At (D) is hole feeding DT78, with a distinct outer rim on Cladophlebis asiatica (CNU-FERN-DHG-2008118). Surface feeding DT103 occurs at (E), delimited by leaf veins, with slot-like damage indicated by arrows on Anomozamites sinensis (CNU- BEN-DHG-2016302). Oviposition DT226 occurs on the midrib of Anomozamites angulatus at (F) (CNU- BEN-DHG-2009567c/p). At (G) is en echelon margin feeding DT81 on Nilssoniopteris longifolius (CNU- BEN-DHG-2010481). A narrow oviposition lesion of DT76 occurs on the midrib of Nilssoniopteris longifolius (CNU-BEN-DHG-2010058) at (H). At (J) is a linear oviposition lesion DT285 and margin feeding DT12 on Elatocladus sp.2 (CNU-CON-DHG-2005518), whose detail is shown at (I). Numerous galls of DT32 occur at (K) consisting of densely distributed, domed circular galls on Sphenobaiera spectabilis (CNU-GIN-DHG-2009546c/p), the scars enlarged at (L). At (M) is linear mine DT411, constrained by parallel leaf veins, on Pterophyllum mentougouensis (CNU-BEN-DHG-2010261). At (N) is gall DT145, with a center chamber on Sagenopteris phillipsii (CNU-CAY-DHG-20211233-2). Galling DT106 occurs on Anomozamites sinensis (CNU-BEN-DHG-2008661) at (O). Short trichomes cover the base of a Ginkgo yimaensis leaf (CNU-GIN-DHG-2006701c/p) at (P), indicated by arrows. Scale bars: white, 10 mm; black, 1 mm. Photography employed a Nikon SMZ18 stereomicroscope. Microphotographic images of mines were taken with a Nikon SMZ18 dissecting microscope (Nikon, Tokyo, Japan), connected to a Leica DFC500 camera with a Nikon DS-Ri-2 digital camera system (Leica, Wetzlar, Germany). All figures were re- organized in Adobe illustration CC. 2022 graphics software (Adobe Inc., San Jose, California, USA). All plant pictures provided by author or modified from a previous related publication (1). Figure S2. Representative insect richness from mid-Mesozoic plant assemblages at Dawangzhangzi, northeastern China, and Rose Creek, Nebraska, USA Insect damage richness from the Early Cretaceous, gymnosperm-dominated plant assemblage at Dawangzhangzi, China (A to N above), and the angiosperm-dominated plant assemblage at Rose Creek, USA (O to Z below). Dawangzhangzi damage. Margin feeding of DT270 on Sphenobaiera sp. (CNU-GIN-DW-2018846) is exhibited at (A); the detail of transverse incisions into two petioles is enlarged at (B). Lunate margin feeding DT357 on Solenites murrayana (CZE-2019040) at (C), An oviposition DT337 lesion, with a clear double ridged outer rim on Liaoningocladus boii (CON-2018003) at (D). Oviposition DT175 with narrow, linear lesions on Baiera valida (GIN-2018256) at (E), enlarged in (F) and (G). Leaf mine DT280, a blotch-like expansion and margin feeding DT13 are on Lindleycladus lanceolatus (CON-2018964) at (H), with detail of an expanded mine at (J), showing a discrete early mine trajectory, after which there is a wide, blotch-like expansion truncated at (I) and (J). Oviposition DT255, with paired, side-by-side lesions, and a DT175 end- to-end oviposition lesion occurring on L. boii (CON-2018354) at (K), of which one of the DT255 lesions is enlarged at (L). Oviposition DT345, consisting of large and ovate lesions on a L. boii stem (CON-2019496) is at (M). Linear leaf mine DT280, with occasional microcoprolites within L. boii (CON-2018043) at (N). Rose Creek damage. Sinusoidal leaf mine DT234, with a large terminal chamber on Pandemophyllum attenuatum (UF-16196-1) is at (O). Leaf mine DT234, showing two adjacent mines, on Pabiania variloba (UF-12708p/c) is at (P). Hole feeding DT2, DT3 and DT5, skeletonization DT61, and a pathogen DT381 necrosis occurs on Pabiana variloba (UF-16215) at (Q). Hole feeding (DT1, DT3), margin feeding (DT12), and fungal necrosis of DT387 on Pabiana variloba (UF-16191) is at (R). Ovate oviposition DT228, with more than ten lesions in a linear array occurring on the midrib of Pandemophyllum kvacekii (UF-16221-3) at (S). Three leaf mines of DT234 with distinctive late instar frass coprolites on Crassidenticulum decurrens (UF-5400) is at (T). Leaf mine DT41, highly winding, with a terminal chamber bereft of frass occurs on Pabiana variloba (UF-16195) at (U). Gall DT386, golfball-like, with a clear single chamber is present on C. decurrens (UF-16131) at (V). Pathogens DT381 and DT388 near the leaf veins of Pandemophyllum kvacekii (UF-16123-2) occur at (W). Gall DT119, with a thick outer wall and a rounded, encompassing center chamber displays six or seven examples occurring in a linear fashion on Crassidenticulum decurrens (UF- 12698) at (X). Gall DT398, with a circular center chamber and wispy extensions into adjacent veins of Pandemophyllum attenuatum (UF-12722-1) is present at (Y) and enlarged at (Z). Scale bars: white, 10 mm; empty 5 mm; black, 1 mm. All plant pictures provided by author or modified from a previous related publications (2, 3, 4). See caption to Figure S1 above for details of photography and production of this figure. Figure S3. Time series analysis of pairwise damage-type comparisons of 134 fossil and three modern plant assemblages partitioned into nine time-intervals A time series analyses (this figure), rather than a pooled analysis (Figure 2 of main text) of the 134 plant assemblages from the 305-million-year interval from the latest Pennsylvanian to the present. Each geochronologic interval is represented by a ternary diagram that contains multiple plant assemblages, except for the Middle Jurassic Daohugou, earlier Early Cretaceous Dawangzhangzi, and later Early Cretaceous Rose Creek deposits (Figures S3C–E), which represent three, separate, mid-Mesozoic plant assemblages (Table S2). All plant assemblages were “standardized” by analyzing the raw data, as “mild lumping”, or “full lumping” categories. (see Appendix S10 for additional details) Figure S4. The most herbivorized plant hosts from five plant assemblages ranging from the Late Pennsylvanian to Early– Late Cretaceous boundary. (A), The Williamson Drive plant assemblage from the Late Pennsylvanian (Gzhelian) of Texas, U.S.A., with 17 DTs on Macroneuropteris scheuchzeri (Medullosaceae) (5); (B), the Aasvoĕlberg-411 plant assemblage from the Late Triassic (Carnian) of South Africa with 28 DTs on Heidiphyllum elongatum (Voltziaceae) (6); (C), the Daohugou plant assemblage from the Middle Jurassic (Callovian) of Inner Mongolia, China, with 37 DTs on Anomozamites villosus (Williamsoniaceae) (1); (D), the Dawangzhangzi plant assemblage from the Early Cretaceous (Barremian) of Liaoning, China, with 59 DTs on Liaoningocladus boii (unknown affiliation) (2); and (E), the Rose Creek plant assemblage from the Early Cretaceous (Albian) of Nebraska, US.A., with 54 DTs on Crassidenticulum decurrens (Lauraceae) (3, 4). Table S1. Plant-assemblage data for beta analysis, minus the quantitative data matrix for functional feeding groups and damage types1. For functional feeding group and damage type quantitative data, see the separate excel file. Absolute Number Herbivorized Plant Abbre- Number Country Age Period Date of specimens FFGs Sources Assemblage1 viation1 of DTs3 (Ma) Leaves (%)2 La Selva LS Costa Rica Recent Quaternary Present 3205 85.6 70 9 (7) Harvard Forest HF USA Recent Quaternary Present 4115 89.0 70 9 (7) Smithsonian Environmental SE USA Recent Quaternary Present 3623 80.7 76 9 (7) Research Center Longmen LM China Gelasian Quaternary 2.15 1028 58.9 18 6 (8) Kimin KIM India Gelasian Quaternary 2.15 130 45 58(freq) 6 (9) Bernasso BEN France Gelasian Quaternary 2.06 535 35 40 8 (10) Berga BEG Germany Piacenzian Neogene 2.80 534 25.029 25 7 (10) Willershausen WSN Germany Piacenzian Neogene 2.80 7932 50 83 7 (10) Rajdanda RAJ India Zanclean Neogene 4.45 --- 400 (herb) --- 6 (11) Subansiri SUB India Zanclean Neogene 4.45 70 39 27(freq) 6 (9) Palo Pintado PAP Argentina Messinian Neogene 6.25 856 33.64 35 7 (12) Iceland 7-6Ma IL7-6 Iceland Messinian Neogene 6.5 415 29.16 20 6 (13) Bnagmai BNA China Tortonian Neogene 8.45 1103 21 36 6 (14) Iceland 9-8Ma IL9-8 Iceland Tortonian Neogene 8.5 409 20.05 18 6 (13) Skarðsströnd- SMR Iceland Tortonian Neogene 9.35 498 18.35 21 6 (13) Mókollsdalur Langhian- LGZ Iceland Tortonian Neogene 9.40 4349 15.63 49 8 (13) Zanclean Iceland-10Ma IL10 Iceland Tortonian Neogene 10 330 8.18 12 5 (13) Pitsidia PIT Greece Tortonian Neogene 10.50 2500 --- --- 6 (15) Tröllatunga- TRG Iceland Tortonian Neogene 10.00 1095 921 17 8 (13) Gautshamar Arunachal ARU India Tortonian Neogene 10.65 150 35 53(freq) 6 (9) Pradesh Kaikorai KAK New Zealand Tortonian Neogene 11.20 --- 35 (herb) 15 4 (16) Brjánslækur-Seljá BRS Iceland Serravallian Neogene 12.00 1612 21.31 12 6 (13) Iceland 12Ma IL12 Iceland Serravallian Neogene 12.00 436 24.54 18 6 (13) Double DOU New Zealand Serravallian Neogene 12.40 --- 127 (herb) 37 6 (16) Huaitoutala HTTL China Serravallian Neogene 12.70 479 70 36 8 (17) Darjeeling DAR India Serravallian Neogene 13.80 137 38 31 5 (18) San José SAJ Argentina Serravallian Neogene 13.80 384 3.9 9 7 (12) Fuotan FT China Langhian Neogene 14.70 3420 40.3 85 8 (19) Zhangpu ZP China Burdigalian Neogene 16.00 1319 --- 70 8 (20) Toupi TP China Burdigalian Neogene 16.90 262 9.5 24 8 (21) Bogotá BOG Colombia Burdigalian Neogene 18.40 955 69.5 90 9 (22) Güvern GVN Türkiye Burdigalian Neogene 18.70 5000 36 41 8 (23) Hindon HID New Zealand Burdigalian Neogene 18.80 584 73 87 8 (24) DSH-Bílina DSH Czech Burdigalian Neogene 19.55 2233 23.16 54 8 (25) Republic LCH Flora- LCH Czech Burdigalian Neogene 19.55 1260 15.58 33 8 (25) Břešťany Republic Mush MUS Ethiopia Aquitanian Neogene 21.73 2200 31.6 35 9 (26) Rott ROT Germany Chattian Neogene 23.08 2326 17.76 52 8 (27) Siebengebirge SIE Germany Chattian Neogene 24.25 2476 19.7 59 6 (27) Enspel ESP Germany Chattian Paleogene 24.80 1622 40.9 39 8 (28) Quegstein QGN Germany Chattian Paleogene 26.63 404 12.6 12 5 (27) Chilga CLG Ethiopia Chattian Paleogene 27.23 1063 26.81 41 7 (28) La Val3 fossil LV3 Spain Rupelian Paleogene 28.10 1337 33.9 28 7 (30) Markam Basin MK-1 China Rupelian Paleogene 33.40 599 24.04 17 5 (31) MK-1 Markam Basin MK-3 China Priabonian Paleogene 34.60 2428 38.39 41 7 (31) MK-3 Renardodden REN Norway Rupelian Paleogene 33.90 413 17.6 18 6 (32) Profen-Süd LC PRO Germany Priabonian Paleogene 37.55 3500 6.32 27 9 (33) Leipzig LEE Germany Priabonian Paleogene 37.55 1558 6.1 22 7 (34) Embayment Aspelintoppen ASP Norway Bartonian Paleogene 40.85 479 16.51 21 6 (32) Eckfeld ECK Germany Lutetian Paleogene 44.30 6748 11 72 8 (35) Bonanza BON USA Lutetian Paleogene 47.30 894 25.8 26 6 (36) Messel MES Germany Lutetian Paleogene 47.80 9334 21 98 7 (35) Wind River Edge WRE USA Ypresian Paleogene 49.13 1369 9.5 31 7 (37) Republic PEP USA Ypresian Paleogene 49.40 1019 49.4 34 9 (38) Laguna del Hunco LH6 Argentina Ypresian Paleogene 51.25 518 29.54 30 7 (39) 6 Laguna del Hunco LH13 Argentina Ypresian Paleogene 51.47 779 29.78 32 9 (39) 13 Laguna del Hunco LH2 Argentina Ypresian Paleogene 51.77 593 33.05 35 9 (39) 2 Laguna del Hunco LH4 Argentina Ypresian Paleogene 51.91 1232 13.39 34 9 (39) 4 Wind River WRI Argentina Ypresian Paleogene 52.42 2519 45.4 39 7 (37) Interior Bighorn BIH USA Ypresian Paleogene 52.65 3612 56 26 7 (37) Fifteenmile Creek FIC USA Ypresian Paleogene 52.70 1821 58 50 7 (40) PN PN USA Ypresian Paleogene 53.40 693 39.11 28 8 (40) Sourdough SOD USA Ypresian Paleogene 53.50 792 32.45 23 7 (41) Cool Period COP USA Ypresian Paleogene 54.20 491 33.3 21 7 (40) Elk Creek ELC USA Ypresian Paleogene 55.20 1008 53.47 28 7 (42) Hubble Bubble HUB USA Ypresian Paleogene 55.80 994 70 38 7 (42) HannaBasin_E HBE USA Ypresian Paleogene 56.00 326 21 19 5 (43) Daiye Spa DAS USA Ypresian Paleogene 56.10 843 37.8 35 8 (42) Dead Platypus DEP USA Ypresian Paleogene 56.40 1016 41.4 28 7 (40) Clarkforkian CLK USA Ypresian Paleogene 56.50 716 29.6 27 7 (41) Lur’d Leaves LUL USA Thanetian Paleogene 58.00 1364 15.01 26 8 (41) Cerrejón CER Colombia Thanetian Paleogene 58.00 507 50.1 28 8 (44) HannaBasin_C HBC USA Thanetian Paleogene 59.00 311 25 20 7 (43) HannaBasin_D HBD USA Thanetian Paleogene 57.50 56 12 5 2 (43) HannaBasin_B HBB USA Thanetian Paleogene 59.00 452 19 8 2 (43) HannaBasin_A HBA USA Thanetian Paleogene 59.00 316 16 9 5 (43) Skeleton Coast SKC USA Thanetian Paleogene 59.39 840 34.6 19 6 (41) Kevin’s Jerky KEJ USA Selandian Paleogene 59.40 1423 398 25 8 (41) Haz-Mat HAM USA Selandian Paleogene 59.40 757 27.9 18 7 (41) Persites Paradise PEP USA Selandian Paleogene 59.40 963 33.5 22 5 (41) Menat MEN France Selandian Paleogene 60.05 938 24-58 35 6 (45) Firkanten FKN Norway Selandian Paleogene 60.85 629 21.9 24 6 (32) Las Flores LAF Argentina Danian Paleocene 62.37 563 59.86 39 8 (46, 47) Peñas Coloradas PEC Argentina Danian Paleogene 62.50 568 59.7 40 8 (47) Salamanca PAL-2 Argentina Danian Paleogene 64.08 1121 64.4 40 9 (47) (Palacio-2) Castle Rock CAR USA Danian Paleogene 64.40 2668 6.74 23 6 (41) Mexican Hat MEH USA Danian Paleogene 65.00 2220 779 32 8 (48) Salamanca PAL-1 Argentina Danian Paleogene 65.22 1082 51.3 32 6 (47) (Palacio-1) Pyramid Butte PYB USA Danian Paleogene 66.00 655 20.6 17 6 (41, 49) Battleship BAT USA Maastrichtian Cretaceous 66.10 461 31.89 31 7 (41, 49) Dean Street DES USA Maastrichtian Paleogene 66.20 524 50.2 32 8 (41) Lefipán LEP Argentina Maastrichtian Cretaceous 66.50 3155 61.2 40 8 (46– 48,50) Somebody’s SBG USA Maastrichtian Cretaceous 66.60 1528 37.2 32 8 (41,51) Garden Luten's 4H LUH USA Maastrichtian Cretaceous 66.80 360 30.56 26 8 (41) Hadrosaur Level Kaiparowits KAP USA Campanian Cretaceous 75.55 1564 38.74 40 8 (52, 53) Puy-Puy PUY France Cenomanian Cretaceous 95.00 1605 22.24 70 9 (54) Rose Creek RC USA Albian Cretaceous 103.00 2084 45.35 114 11 (3, 4) Dawangzhangzi DW China Aptian Cretaceous 125.00 2176 34.01 65 9 (2) Daohugou DHG China Callovian Jurassic 165.00 2001 49.45 76 11 (1) El Pedregal ELP Spain Aalenian Jurassic 174.00 14 --- 11 4 (55) Zhenzhuchong ZZC China Hettangian Jurassic 199.30 127 48 21 6 (56) Xujiahe XJH China Rhaetian Triassic 201.00 469 42.2 29 7 (56) Yangcaogou YCG China Rhaetian Triassic 205.00 220 --- 11 6 (57) Yipinglang YPL China Norian Triassic 214.00 2081 8 10 5 (58, 59) Birds River 1114 Bir 111 South Africa Carnian Triassic 232.50 15503 15.33 41 10 (6) Greenvale 1214 Gre 121 South Africa Carnian Triassic 232.50 2966 2.06 11 6 (6) Boesmanshkoek Boe 111-C South Africa Carnian Triassic 233.50 700 2.71 6 3 (6) 111C4 Boesmanshkoek Boe 112 South Africa Carnian Triassic 232.50 1197 0.5 2 0 (6) 1124 Cyphergat 111A4 Cyp 111-A South Africa Carnian Triassic 232.50 6377 2.62 27 8 (6) Kannaskop 1124 Kan 111 South Africa Carnian Triassic 232.50 1538 2.93 9 4 (6) Kannaskop 1114 Kan 112 South Africa Carnian Triassic 232.50 2387 1.68 11 6 (6) Telemachus Tel 111 South Africa Carnian Triassic 232.50 6681 1.42 16 8 (6) Spruit 1114 Kommandantskop Kom 111 South Africa Carnian Triassic 232.50 1213 1.65 10 6 (6) 1114 Vineyard 1114 Vin 111 South Africa Carnian Triassic 232.50 2217 2.89 10 6 (6) Elandspruit 1114 Ela 111 South Africa Carnian Triassic 232.50 1154 4.24 10 6 (6) Elandspruit Ela 112-A South Africa Carnian Triassic 232.50 1295 5.79 6 5 (6) 112A4 Kraai River 3114 Kra 311 South Africa Carnian Triassic 232.50 1387 3.03 6 4 (6) Kraai River Krb 111 South Africa Carnian Triassic 232.50 2006 9.38 3 3 (6) Bridge 1114 Lutherskop 5114 Lut 511 South Africa Carnian Triassic 232.50 634 4.1 11 5 (6) Lutherskop 41124 Lut 4112 South Africa Carnian Triassic 232.50 744 1.75 6 4 (6) Lutherskop 3114 Lut 311 South Africa Carnian Triassic 232.50 5784 3.48 25 9 (6) Waldeck 1114 Wal 111 South Africa Carnian Triassic 232.50 1695 5.25 12 5 (6) Konings Kroon Kon 223 South Africa Carnian Triassic 232.50 517 0.77 4 2 (6) 2234 Konings Kroon Kon 222 South Africa Carnian Triassic 232.50 2973 1.01 14 8 (6) 2224 Konings Kroon Kon 211- South Africa Carnian Triassic 232.50 774 4.01 10 5 (6) 211A4 & 2214 A & 221 Konings Kroon Kon 111- South Africa Carnian Triassic 232.50 1190 1.51 5 2 (6) 111A4 A Konings Kroon Kon 111-C South Africa Carnian Triassic 232.50 573 4.71 11 6 (6) 111C4 Peninsula 3214 Pen 311 South Africa Carnian Triassic 232.50 2315 1.51 7 4 (6) Peninsula 4214 Pen 421 South Africa Carnian Triassic 232.50 870 2.41 13 7 (6) Peninsula 3114 Pen 311 South Africa Carnian Triassic 232.50 1785 2.07 13 7 (6) Peninsula 4114 Pen 411 South Africa Carnian Triassic 232.50 6807 3.35 19 8 (6) Klein Hoek 111B4 Kle 111-B South Africa Carnian Triassic 232.50 1267 3 9 6 (6) Klein Hoek 111C4 Kle 111-C South Africa Carnian Triassic 232.50 2930 3.04 17 7 (6) Kapokkraal 1114 Kap 111 South Africa Carnian Triassic 232.50 1965 17.44 28 9 (6) Nuwejaarspruit Nuw 111- South Africa Carnian Triassic 232.50 546 6.06 7 4 (6) 111A4 A Nuwejaarspruit Nuw 111- South Africa Carnian Triassic 232.50 2188 3.43 21 7 (6) 111B4 B Nuwejaarspruit Nuw 211 South Africa Carnian Triassic 232.50 763 10.22 13 6 (6) 2114 Winnaarspruit Win 111 South Africa Carnian Triassic 232.50 1679 4.71 17 7 (6) 1114 Morija 111A4 Mor 111A South Africa Carnian Triassic 232.50 539 5.94 4 2 (6) Morija 111B4 Mor 111B South Africa Carnian Triassic 232.50 1628 3.56 7 2 (6) Makoaneng 1114 Mak 111 South Africa Carnian Triassic 232.50 1308 1.45 7 4 (6) Mazenod 1114 Maz 111 South Africa Carnian Triassic 232.50 1014 6.05 11 5 (6) Mazenod 2114 Maz 211 South Africa Carnian Triassic 232.50 2279 9.26 32 9 (6) Hlatimbe Valley Hla 111 South Africa Carnian Triassic 232.50 515 100 0 0 (6) 1114 Hlatimbe Valley Hla 212 South Africa Carnian Triassic 232.50 723 3.04 11 6 (6) 2124 Hlatimbe Valley Hla 213 South Africa Carnian Triassic 232.50 1943 3.04 18 8 (6) 2134 Umkomaas 1114 Umk 111 South Africa Carnian Triassic 232.50 12788 4.46 38 10 (6) Sani Pass 1114 San 111 South Africa Carnian Triassic 232.50 1340 2.09 11 7 (6) Qachasnek 1114 Qac 111 South Africa Carnian Triassic 232.50 2130 0.61 7 6 (6) Matatiele 1114 Mat 111 South Africa Carnian Triassic 232.50 6343 3.45 25 10 (6) Golden Gate 1114 Gol 111 South Africa Carnian Triassic 232.50 1326 8.07 17 7 (6) Little Switzerland Lis 111 South Africa Carnian Triassic 232.50 9912 7.07 20 9 (6) 1114 Aasvoëlberg 1114 Aas 111 South Africa Carnian Triassic 232.50 3308 1.03 12 6 (6) Aasvoëlberg 2114 Aas 211 South Africa Carnian Triassic 232.50 2061 4.85 12 7 (6) Aasvoëlberg 3114 Aas 311 South Africa Carnian Triassic 232.50 11677 10.15 20 8 (6) Aasvoëlberg 411 Aas 411 South Africa Carnian Triassic 232.50 20358 4.15 44 11 (6) Askeaton 111 Ask 111 South Africa Carnian Triassic 233.50 1061 4.15 8 5 (6) Monte Agnello MOA Italy Anisian Triassic 244.60 649 12.13 20 8 (51) Kühwiesenkopf KHF Italy Anisian Triassic 244.60 1260 11.6 26 9 (51) Kayitou KYT China Changhsingian Permian 253.00 1712 6.48 25 7 (60) Clouston Farm CLF South Africa Wuchiapingian Permian 256.00 1828 1.4 22 7 (61) Bletterbach BLE Italy Wuchiapingian Permian 257.00 1531 2.2 17 10 (51) Hammanskraal Ham 111 South Africa Capitanian Permian 261.90 15240 9.25 36 8 (62) 111 South Ash Pasture SAP USA Wordian Permian 266.00 756 4.1 22 6 (63) La Golondrina LAG Argentina Wordian Permian 266.00 2523 7.37 35 7 (64) Tregiovo TRE Italy Kungurian Permian 275.50 464 3.24 5 4 (51) Colwell Creek CCP USA Kungurian Permian 275.50 2140 61.58 51 9 (65) Pond Quitéria QUI Brazil Kungurian Permian 281.00 27 7.41 --- --- (66) Faxinal FAX Brazil Kungurian Permian 281.00 171 7.02 --- --- (67) Mitchell Creek MCF USA Artinskian Permian 284.50 820 5.9 22 7 (68) Flats Taint TAI USA Artinskian Permian 284.50 1346 --- --- --- (69) Wuda WUDA China Artinskian Permian 285.50 10000 0.4 21 8 (70) Morro do Papaleo MDP4 Brazil Artinskian Permian 287.50 16 12.5 --- --- (67) (Stratum 4) Morro do Papaleo MDP7/8 Brazil Artinskian Permian 287.50 138 9.42 --- --- (67) (Stratum 7/8) Paraná Basin PAB Brazil Artinskian Permian 289.00 850 8 --- 4 (67) Ranigani Barakar RAB India Artinskian Permian 289.00 --- --- --- 4 (71) Río Genoa Fm. RGE Argentina Artinskian Permian 289.00 291 0.27 12 --- (72) Coprolite Bone CBB USA Sakmarian Permian 293.00 598 18.27 11 4 (73) Bed Sanzenbacher SR USA Asselian Permian 296.00 1390 --- 48 9 (74) Ranch Williamson Drive WD USA Gzhelian Pennsyl- 301.50 1830 23.85 46 5 (5) vanian Kinney Brick KIN USA Kasimovian Pennsyl- 305.00 2254 1.6 9 4 (75) Quarry vanian Notes 1. Color highlights represent the 131 fossil and three modern plant assemblages that were used for further analyses in this study. Yellow shading represents Cenozoic plant assemblages; blue shading represents Mesozoic plant assemblages; and green shading represents Paleozoic plant assemblages used in the beta analysis study. 2. Dashes represent the absence of values in the original published datasets. The designation (freq) represents the DT frequency, and (herb) are the herbivorized specimens in the database; more detailed results that can be found in associated references. 3. Due to the updated version for the old version Damage Guide (76), there have been minor changes in DT assignments, such as removal of a DT from one functional feeding group or a reassignment to another functional feeding group. Nevertheless, the data listed here show minimal difference with the results from the raw data, details of which are in the online excel file of Paleozoic to Cenozoic datasets. 4. Basic detailed plant–insect associational data for these Late Triassic Molteno plant assemblages cited in reference (6). Table S2. Average values of turnover and nestedness for plant assemblages from time intervals (related to Figure S3 above) Note: N represents the number of plant assemblages that met the standards for analysis examined for each time interval. The T/N Ratio is Turnover divided by Nestedness, reflecting the relationship of these metrics (Table S3). Table S3. General definitions and descriptions of turnover and nestedness of Beta diversity for the β-analysis studies Metric Notation Definition Application and meaning The difference in species The same number of DTs shared in two or more composition between two host-plant species or plant assemblages. For localities or communities example, when two host plants share fewer of the Turnover β(turnover) that likely reflect the pre- same DTs, the turnover is higher; conversely, sence and absence of rare when they share more of the same DTs, the species in the assemblages turnover is lower The demonstration that The diversity of DTs on host plants that form a species assemblages in distinct pattern. When the similarity of the diverse Nestedness β(nestedness) species-poor sites are a associations between DTs and host plants is higher, subset of the assemblages in nestedness is stronger; conversely, the lower the more species-rich localities similarity, nestedness is weaker. The similarity/dissimilarity of DTs and plant hosts The similarity among among different plant assemblages. The higher the Similarity 1-β localities value indicates the more similar (or lower diversity) there is between two assemblages turnover < nestedness Total insect herbivory is mainly determined by nestedness Higher than 1: turnover contributes more to insect herbivory; turnover / nestedness (T/N) lower than 1: nestedness contributes more to insect herbivory. Appendix S1. Hypotheses explaining angiosperm expansion in the Early Cretaceous Features favoring angiosperm ecological expansion include establishment of a positive feedback cycle resulting in greater growth rates from an increased nutrient supply. This nutrient supply was supported by more rapid plant uptake from easily decomposed litter (77), efficient harvesting of light from large, planated leaves (78,79), shortening of the reproductive cycle attributable to faster growth rates, and survivability especially in disturbed habitats (79). Also present were greater efficiencies in metabolizing a greater diversity of chemical defenses such as alkaloids (80,81), accommodation to major climate changes accentuating clade competition (82,83), and a multidriver hypothesis that encompasses several of these individual characteristics (84,85). Appendix S2. A brief history of arthropod and pathogen mediated herbivory from Devonian to recent A steady but fluctuating increase in the richness of damage types (DTs) and functional feeding groups (FFGs) through time (Figure 1) reflects the various modes by which insects consumed plants from the Middle Devonian to the Recent. Our data indicate that there is an absence of plant assemblage data from those time swaths represented by the Late Devonian, Mississippian (Early Carboniferous), Early Triassic, Early Jurassic, and Late Jurassic (86,87). Our account below provides a history of herbivory as encapsulated by four phases of plant–insect interaction expansion (86). Phase 1. The early history of herbivory during the Devonian to Mississippian time interval has been summarized previously (86–90), although data during this phase are sparse. The initial pulse of herbivory was launched among the earliest nonvascular plants during the Early Devonian that consisted of spore consumption, stylate lesions, and borings on vascular plant stems (91). During the Middle Devonian, margin feeding, surface feeding, piercing and sucking, and galling constituted the four FFGs present on well preserved liverwort thalli (92). By the Middle Mississippian (Early Carboniferous), folivory on seed plants was established. Whether this sparse record of interactions represents an intrinsic pattern, or the absence of searching, remains a debated issue (89). Phase 2. A second phase during the Pennsylvanian (Late Carboniferous) to Permian corresponds to an increase in herbivory that was considerably more diverse than that of Phase 1. During the Early Pennsylvanian, an initial pulse of herbivory consisted of piercing and sucking, oviposition, and galling, overwhelmingly on plant axes such as stems (93). This was followed by modest increases of margin feeding and pathogens occurring on foliage and seed predation during the rest of the Pennsylvanian, resulting in an increase to six FFGs. At this time, all plant organs were affected by herbivory, including the pinnules of often large pinnate fronds, seeds, occasionally true roots, rarely indurated tissues such as wood, and lycopod leaf cushions, but continuing into the Pennsylvanian–Permian boundary interval (89,91,94). There was a major shift during the early Permian, with a second surge of herbivory characterized by external feeding and insect oviposition on foliage that increased to nine FFGs, of which skeletonization and borings were added to the previous spectrum (5,64,90). Notably, most of the richness and intensity of herbivory was transferred from free sporing plants to seed plants during this time, initially on medullosans and cordaites, and later onto peltasperms, gigantopterids, and glossopterids in Euramerican, Cathaysian, Gondwanan, and perhaps Angaran floras (51,62,74,88). Toward the end of the Permian, many plant assemblages displayed smaller sized leaves, likely due to increased aridity that decreased herbivory levels (61,62). This was soon followed by a major associational setback induced by the Permo–Triassic ecological crisis, after which it was not until the late Middle Triassic that herbivory levels of the latest Permian were reached (51). Phase 3. Herbivory throughout the Triassic, Jurassic, and into the mid Early Cretaceous, when angiosperms made their initial ecological expansion, constitute Phase 3. Phase 3 was comprised of overwhelmingly of ferns and especially gymnosperms as plant hosts, and new lineages of herbivorous insects, notably early holometabolous lineages (86). A major increase in DT richness occurs during the Late Triassic (Figure 1), reflecting the establishment of new lineages of plants and their herbivores, and associations resulting from the Permian–Triassic (P-Tr) boundary crisis. Such a pattern has been documented at a broader taxonomic scale (86). Nevertheless, Phase 3 was diminished by the major, mid-Cretaceous ecological expansion of angiosperms (and ferns) representing Phase 4 (3,74), albeit some of the Phase 3 biota survives today in habitats such as coniferous forests. After the P-Tr ecological crisis, new plant and herbivore lineages colonized newly formed habitats, expanding to ten FFGs that are recorded on ferns and especially gymnosperms such as voltzialean conifers, cycads, peltasperms, corystosperms, and a diverse array of ginkgoaleans, such as those described from the Late Triassic of South Africa, exemplified by the Aasvoëlberg 411 plant assemblage (6,95). Little is known about herbivory trends in the ensuing Jurassic, which evidently was associated with diversification of new gymnosperm lineages of bennettitalean, ginkgoalean, caytonialean, coniferalean, other gymnosperm lineages, and ferns. The first documentation of herbivory from a Jurassic bulk plant assemblage is the Middle Jurassic Daohugou plant assemblage of Northeastern China (1). Previous studies have examined herbivory trends for certain elements throughout the mid-Mesozoic, such as the broad spectrum of herbivory on coniferaleans (57), oviposition on coniferaleans (96), and mining on bennettitaleans (97). A recent study indicates a much more extensive spectrum of herbivory, representing ten mining DTs (97), more than previously expected. Examination of a forty-million-year-younger gymnosperm-dominated plant assemblage from Dawangzhangzi, of mid Early Cretaceous age, exhibits a diverse array of 65 DTs in nine FFGs, including its most herbivorized plant, the broadleaved coniferalean Liaoningocladus boii, and two less but moderately herbivorized narrowleaved czekanowskialean species of Czekanowskia (2). Notably, the Dawangzhangzi plant assemblage shares about 52% of its DTs in common with the older Daohugou plant assemblage (1). Other plant assemblages such as the Late Triassic Aasvoelberg-411, Middle Jurassic Daohugou, and Early Cretaceous Dawangzhangzi have been extensively examined (1,2,6,97), The overall pattern of herbivory for Phase 3 provides a consistent pattern, as these plant assemblages represent much greater sample sizes than almost all analogous previous studies of Phase 2. Phase 4. The 62 million-year-younger, angiosperm-dominated, Early Cretaceous, Rose Creek plant assemblage of Nebraska, U.S.A., represents a well-documented early stage of initial angiosperm diversification (1–4). Rose Creek exhibits the most diverse feeding damage of any Mesozoic plant assemblage, constituting 114 DTs and the eleven FFGs of hole feeding, margin feeding, skeletonization, surface feeding, oviposition, piercing and sucking, mining, galling, seed predation, borings, and pathogens (3,4). The slightly younger, earliest Late Cretaceous Puy-Puy plant assemblage also have been examined extensively for herbivory (54). Plant assemblages of the Late Cretaceous have been partly examined (53), later undergoing a major associational setback resulting from the Cretaceous–Paleogene (K-Pg) ecological crisis (Figure 1). After this event, herbivory in various continents reflect different time intervals of recovery (38,45,46,48). However, during and after the transient Paleocene–Eocene Thermal Maximum (PETM) at 56 Ma, ten million years after the K-Pg crisis at 66 Ma, herbivore recovery proceeded in a more subdued manner, affected by biotic and abiotic factors caused by environmental change (40,42). This recovery could be attributed to a more diverse suite of host plants and their herbivorous arthropods and pathogens that are more analogous to extant lineages (7); or alternatively, to a more accurate reconstruction of Cenozoic climate based on better preserved, fossil deposits that support more realistic environmental reconstructions (26). In summary, Phase 4 began with diversification of angiosperms (and ferns) in the late Early Cretaceous and their expansion during the Late Cretaceous and later occurrences throughout the Cenozoic after an initial lull immediately after the K-Pg event. Nevertheless, herbivory rebounded with elevated DT richness that increased immediately before and during the PETM and several subsequent thermal events. Summary. Extant herbivory is principally recorded on angiosperms, a feature that, with the exception of coniferous forests at higher latitudes and altitudes, is consistent with mostly angiosperm-dominated landscapes (98) of the late Early Cretaceous. Nevertheless, there have been four major global phases of herbivory, their herbivores, and associations during the past 420 million years (86). The earliest well documented plant assemblages were dominated by spore-bearing nonvascular and vascular plant lineages present during the Devonian, followed by the emergence of new lineages of spore-bearing vascular plants and early seed plants in the following Carboniferous and Permian periods. These assemblages in turn were succeeded by surviving and new lineages of ferns and gymnosperms that colonized new habitats after the end-Permian ecological crisis. Most of the global flora was replaced by the last major floral transition of angiosperms during the mid Cretaceous that continued through time and largely dominate the present-day world (86,88,99). These major transitions in globally distributed floras mirror parallel, albeit complex, developments in the evolution of herbivory (86,88). Much of the attention regarding fossil herbivory has been devoted to this later gymnosperm–angiosperm transition (2–4,87,100). Nevertheless, many earlier plant assemblages and associated herbivory patterns remain unexamined. Appendix S3. The standard view for the emergence of modern-style terrestrial herbivory by arthropods and pathogens The most obvious herbivory trend through time is the monotonic expansion of functional feeding group (FFG) and especially damage type (DT) richness. However, the mode in which this increase occurred, leading to the modern pattern, is poorly known. Various efforts recently have considered the phylogenetic or ecological context in which the modern pattern initially appeared (22,85,101) and have mostly concluded that the current pattern of herbivory began with the emergence of angiosperms during the Cretaceous. Entomological studies centered during the period of 2005–2017 overwhelmingly indicate that diversification of herbivorous insect lineages paralleled the initial diversification of angiosperms during the later Early Cretaceous. These studies include katydids (102), planthoppers and leafhoppers (103), scarab beetles (104), leaf beetles (105), weevils (106), leafmining flies (107), midget moths (108), leafroller moths (109), moths and butterflies in general (110), ants (111), and bees (112). The emergence of the modern style of herbivory is similar to that seen for specific, current interactions with live plant tissues exemplified by ovipositing insects (113), by endophytic insects such as miners and gallers (114), and by some pollinating insect lineages (99,115,116). However, many pollinator lineages display deeper divergences in the Mesozoic associated with gymnosperm hosts that substantially preceded the appearance of angiosperms (76,99,117,118,119). Abiotic factors may have controlled this pattern. Factors such as warm climates, arid environments, biotic pressures from the Mesozoic Lacustrine Revolution, the emergence of parasitoids, or chemical and physical defense measures that induced plant-host switches (87,89,100,120–122). Some of these events often are associated with development of new insect mouthpart structures (123–125). Three types of pertinent investigations have examined the spatiotemporal distribution of modern herbivory. The results of these studies indicate that host-plant diversity is the most pervasive factor in contributing to herbivory (126,127); that latitude also is a good predictor of herbivory (128); and temperature is associated with increased herbivory (129). Evidently, herbivory is directly influenced by multiple factors, which also include tree canopy height, mean annual temperature (MAT), mean annual precipitation (MAP), and the role of specific microorganisms such as nitrogen-fixing root nodules (12,130,131). From these and other studies, a feature in common with these factors is the citing of angiosperm diversification as the central event that gave rise to the modern insect herbivore fauna. For our long-term data, spanning approximately 305 million years and 134 plant assemblages (Table S1), we employed the coarse-grained approach of beta analysis (132,133). By contrast, centered during the largely subsequent period of 2013–2024, entomologists have documented diversification events of insect herbivore lineages on gymnosperms during the Middle and Late Jurassic, considerably antedating the initial Early Cretaceous appearance of angiosperms. These latter studies from 2013 to 2024 include cicadas (134), scale insects (135), planthoppers (136), beetles in general (137), weevils (138), early appearing moths (139,140), common sawflies (141), web-spinning sawflies (142), approximately 52 family-level herbivorous lineages of Orthoptera (grasshoppers, crickets, and allies), Hemiptera (true bugs, hoppers, aphids, cicadas, and allies), Coleoptera (beetles), Lepidoptera (moths), and Hymenoptera (sawflies), based on fossil occurrence data (100). A prominent example is the overwhelming presence of predatory Neuroptera in modern habitats that had different habits during the Middle Jurassic (51,143). Neuroptera of the Daohugou plant assemblage often formed tight associations with plants, including herbivory, pollination, and mimicry behaviors (144–146). Almost all Jurassic neuropteran taxa are extinct; nonetheless, these behaviors are not represented by neuropterans in the present-day world. Above all, the ancestral lineages of insect taxa often have evolved in intricate ways, especially morphological features and behavioral and ecological habits, compared to their modern descendant lineages (145). Appendix S4. An extended explanation of the DT and FFG trends of Figure 1 The trends displayed in Figure 1 are susceptible to influences by the number of examined specimens (147), and the particular dataset that was assembled. (The functional feeding group [FFG] and damage type [DT] data providing the raw data for this figure are not included.) The datasets indicate that the richnesses of Cenozoic plant assemblages likely would be higher than Mesozoic plant assemblages for several reasons (148). First, the rare records of borings and seed predation mostly reflect occasional preservation of wood, trunk, seeds, and other indurated seed-plant organs on compressed fossil slabs. Second, plant damage scars are easily overlooked or difficult to detect on roughened surfaces of specimens such as petioles, rachises, and stems. Third, except for a few seed predation records documented from Permian plant assemblages (65,68), seed predation similarly is rare but more importantly rarely inventoried in studies. Fourth, pathogen DTs also have poor fossil evidence, as it is primary arthropod-induced damage that attracts the attention of researchers, rather than inconspicuous pathogen damage, such as that described from the angiosperm- dominated Rose Creek plant assemblage (3,4). Alternatively, an important distinction between plant assemblages occurring before and after the Early Cretaceous–Late Cretaceous boundary is the richness of oviposition. It seems unlikely that the oviposition FFGs would exhibit a lower richness for the Cenozoic, rather than the late Paleozoic and much of the Mesozoic (Figure 1). An explanation for this shift is that several lineages with medium to (very) large size insects possessed large, slicing ovipositors during the late Paleozoic to earlier Mesozoic, which provided a suite of distinctive, often large, oviposition lesions that did not survive into the Cenozoic (96,113). This pattern is generally explained by regional or global extinction events such as the end-Cretaceous ecological crisis or biotic revolutions such as initial angiosperm diversification. Appendix S5. Time series β-diversity analysis of turnover and nestedness among plant assemblages (related to Figure S3 above) The most revealing analysis documents a time series consisting of 134 plant assemblages that structure a ternary turnover–nestedness–similarity pattern. This pattern is formatted as a beta analysis ternary diagram in which turnover, nestedness, and similarity form the three equilateral sides of a triangle, within which a distinct field of data is plotted (Figure S3; Table S2; Dataset S2). Each ternary diagram has a data field consisting of numerous datapoints from multiple plant assemblages, each datapoint of which indicates the mean DT richness for two host plants within an assemblage. From these data a distinct pattern emerged, consisting of Late Pennsylvanian to Triassic plant assemblages with dispersed datapoints (Figure S3A–B) that are different from tightly clustered datapoints of Middle Jurassic to modern assemblages (Figure S3C– I). This Middle Jurassic–modern pattern is densely packed by datapoints of pairwise comparisons of DTs, except for two Early Cretaceous plant assemblages, where the datapoints are sparse but nevertheless located within the distinctive field established by modern plant assemblages (Figure S3I). Late Paleozoic. Although sparse, Late Pennsylvanian and Permian plant assemblages show low values of turnover of 0.32–0.41 and nestedness of approximately 0.1, which indicate that there are more DTs shared in common but with low subset relationships of DTs on their plant hosts. This pattern indicates that links between DTs and major host plants are dissimilar across the assemblages. Triassic. The trend of highly dispersed values for the three values of turnover, nestedness, and similarity is notable for the Triassic. This pattern reflects a condition of host plants that have very different distributions of shared/not shared DTs, which define very dissimilar, nested, DT–host plant subsets of the larger, encompassing sets. Middle Jurassic–Early Cretaceous. The Middle Jurassic plot expresses low ratio turnover values of 0.05– 0.45 (average 0.14), whereas nestedness values display a wider threshold of 0.02–0.9 (average 0.42). This trend for the Daohugou plant assemblage also is present for the Early Cretaceous Dawangzhangzi plant assemblage, though muted, wherein it displays a similar low turnover value of 0.02–0.45 and a high interval range of values of 0.16–0.83 for nestedness. The angiosperm-dominated plant assemblage at Rose Creek is characterized by low pairwise turnover and nestedness range of insect DTs of 0.15–0.4. Collective data from these three, highly sampled, plant assemblages are consistent with the data field established by the modern herbivory pattern outlined in Figure S3I and other more recent plant assemblages. The pattern exhibited by the three mid-Mesozoic plant assemblages of Daohugou, Dawangzhangzi, and Rose Creek–show that a greater number of DTs were shared on different host plants that collectively represent a distinctive structure of links between DTs and their host plants. This pattern defines a common herbivory pattern dominating these and more recent plant assemblages. Late Cretaceous–Modern. The Late Cretaceous, Paleogene, Neogene, and modern ternary plots show a stable distribution of values that was established in the mid Mesozoic. The major clusters display a wide threshold for turnover values of 0.1–0.6, and nestedness values of 0.1–0.95, with mean values of 0.21 and 0.39, respectively (Table S2). The modern data show that the mean values of turnover at 0.13, and nestedness at 0.45 is very close to the mean values of the Middle Jurassic plant assemblage. The pattern of low turnover and a wide threshold for nestedness illustrates that herbivory continues in a similar pattern for this broad time interval, with co-occurrences consisting approximately of the same DTs on host plants. This more recent herbivory pattern conservatively appeared no later than late Middle Jurassic. Appendix S6. Consequences of the NMDS study Studies documenting terrestrial herbivory through time have been examined at short geological time-scales and typically have focused on key geologic events such as the end-Cretaceous ecological crises (45,46,48,49), and the Paleocene–Eocene Thermal Maximum (6,35,42) that involve time intervals during angiosperm dominance (148–150). Larger-scale studies extending to the Mesozoic (3,4,6,53), or late Paleozoic (5), are rare due to a paucity of outcrops, absence of sufficient preservation, or lack of investigator interest. Nonetheless, our results indicate that DT and FFG richness of plant assemblages during the Middle Jurassic, Early Cretaceous, and earliest Late Cretaceous are highly similar to results from short to medium term time- scale studies of Cenozoic and modern plant assemblages (Figures 1, 6; Appendices S2, S4). For modern in situ studies, some robust, analogous studies typically use FFG-DT methods for assessing insect herbivory (149), whereas other studies do not (151). The relationships among plant clades/groups, FFGs, and plant assemblages, as indicated by NMDS (Figure 4), display a general consistency with previous studies, such as the late Paleozoic of north-central Texas (63,68,74), mid Mesozoic of northeastern China (2–4), and Cenozoic of the USA Western Interior (26). Our global, late Paleozoic to modern analysis (Figure 4) also validated associational patterns from previous studies and modern plant assemblages that overlapped between the ellipses of the Cenozoic and Mesozoic but did not overlap with ellipses of the late Paleozoic and many scattered Triassic plant assemblages. This absence of an overlap largely indicates that the present insect herbivory pattern has a close relationship with later Mesozoic and Cenozoic plant–arthropod associations but is distant from the late Paleozoic and almost all of the Triassic. Such a disparate pattern may be explained by more recent insect lineages associating with a broader range of FFGs. An alternative explanation is that more recent insect herbivores have narrower DT functional breadths and become restricted to plant hosts within a single taxonomic group whereas earlier herbivores exist with broader DT functional breadths that were widely distributed on multiple, unrelated, plant taxonomic groups (26,126). The confined ellipse reflecting modern data (Figure 4B) is highly concentrated (150) compared to that of fossil data, suggesting a high level of plant assemblage integration. Appendix S7. Major leaf morphotypes and leaf mass per area Major leaf morphotypes consist of three groups: ferns, gymnosperms (broadleaved and narrowleaved), and angiosperms. Reproductive structures, such as fruits and seeds, and certain vegetative structures including stems, branches, unidentified shoots, and unassigned foliage were not included as data. Later Middle Jurassic gymnosperm vegetation is mostly dominated by gymnospermous plants such as Anomozamites that developed broadleaved fronds with pinnular parallel venation and a network of interspersed delicate veins, similar to co-occurring broadleaved leaves such as broadleaved Sagenopteris and narrowleaved Sphenobaiera, Solenites, and Yanliaoa (1,152). The association between a leaf taxon and DT richness was analyzed in this study. Data on the shape, size, lifespan, and leaf mass per area were often collected in the same datasets that record plant specimen identifications and DT data (42,153,154). However, such data were not available for all 180 initial plant assemblages due to the vagaries of fossil preservation and general investigator practice. Nevertheless, some of these traits, such as leaf mass per area, can be a useful index for estimating mean annual temperature (MAT), mean annual precipitation (MAP), and other aspects of the paleoenvironment (155). However, recording leaf mass per area requires excellent preservation of the leaf petiole and margin (154) – features that are inapplicable for fern pinnules, scale leaves, and conifers with needleleaved foliage (154,156). For these reasons, paleotemperature estimates based on leaf mass per area were not suitable for this study. This is especially true for the interval of the fossil record lacking dicotyledonous angiosperm leaves. Consequently, the relative analysis of paleoenvironmental elements in deep time was based on results reported from previous research (51,56). Appendix S8. Cenozoic patterns resulting from damage-type functional breadth The extent of the distribution of a DT on its plant host or hosts is its DT functional breadth (87). It turns out that the proportion of DTs with narrow DT functional breadth (“specialized”) in Cenozoic plant assemblages is considerably less than that of Mesozoic and late Paleozoic plant assemblages (Figure 5). Data for DT functional breadth is broad (“generalized”) for Cenozoic plant assemblages (22,37,52,150,151). The low percentage of DTs with narrow DT functional breadth in the Cenozoic from our study is consistent with the general trend from the Cretaceous to the Recent whereby narrow DT functional breadth contributed to about 20 DTs and approximately 10% of the frequency of damage (151), a pattern that increases during the Paleocene–Eocene Thermal Maximum (42). The lower proportion of narrow DT functional breadth in Cenozoic plant assemblages probably is the result of the greater presence of plants retaining older associations on multiple plant hosts and less the possibility of newly formed herbivore associations on multiple hosts. This explanation would imply that a more complex herbivore pattern likely developed during the later Mesozoic (Figure 5) than previously understood. Notably, these relatively granular results from beta diversity analyses can uncover general patterns that typically are submerged in the minutiae of fossil data. This is especially so, as herbivore damage occur in numerous combinations, at highly variable intensities, and as diverse modes in plants, with or without the advantages of physical and chemical antiherbivore defenses (3,123,148,149). Appendix S9. Herbivory-related features of angiosperms present in earlier gymnosperm plant assemblages Many of the biological features occurring in angiosperm-dominated plant assemblages of the Cretaceous to the present also occurred in the earlier, Middle Jurassic Daohugou plant assemblage. Five items indicate these equivalencies: growth-form diversity of the plant community, co-associations between host plants and their insect pollinators, presence of diverse chemical and physical defenses to deter herbivory, plant–insect mimicry, and the narrowness, or “specialization” of DT functional breadth. Growth-form diversity. The Middle Jurassic Daohugou plant assemblage consisted of a ground cover of lichens, moss, and liverworts, an understory principally of ferns, a middle stratum of mostly gymnospermous shrubs, and a canopy of taller arborescent plants such as ginkgophytes (157–159) (Figure 6). Many of these arborescent plants became extinct during Early Cretaceous (100,160,161) but the growth- form structure of this Jurassic community is analogous to many plant communities of today. Pollination mutualisms. Although insect pollinators were exclusively on gymnosperms during the Jurassic and Early Cretaceous, the structure of the pollinator communities were modern in several aspects (162). Several major pollination modes of fluid-feeding were present: long-proboscid insects representing four orders of insects, thrips with specialized wing structures for transporting pollen, beetles with mandibulate mouthparts for accessing and consuming pollen, and brachyceran flies with sponging mouthparts for imbibing pollination drops (99,100,116,118,119). Gymnosperm plant hosts equally were diverse in their reproductive structures, ranging from a common cupular tube leading to ovular pollination drops in Caytoniaceae, a salpinx type of tubule in Pentoxylaceae, pappus tubes in Gnetales, and a bizarre combination of a nectary lined funnel–pipe–micropyle system in some Cheirolepidiaceae (117). For bennettitalean Williamsoniaceae, there was access likely to beetles via pollen sacs lining the internal surfaces of pollen organs and nectary-like structures of conspecific ovulate organs (91,117,144). Antiherbivore plant defenses. Antiherbivore chemical and physical defenses were present in Jurassic plant assemblages. Chemical defenses included the resin duct system in many coniferous plants that evidently warded off the borings and foliage feeding of beetles (3). The presence of herbivore deterrent chemicals such as cycasin in cycadaleans and 2-hexenal in ginkgoaleans of extant taxa (163,164) likely extended to ancestors during the preangiospermous Mesozoic. Physical plant defense structures included dense, stiff trichomes along the frond midribs of Anomozamites villosus (164), and shorter defensive trichomes on the leaf blades of Ginkgoites sp. 1 (Figure 6, Figure S1). Plant–insect mimicry. The oldest documented mimicry extends to the Permian, described for the orthopteran Permotettigonia gallica on the presumed seed fern Taeniopteris sp. (165). More extensive examples of mimicry, including camouflage, has been a major focus for the Middle Jurassic Daohugou plant assemblage. Several, distinct examples of mimicry have been established between common plant hosts of lichen, fern, ginkgophyte, bennettitalean, and cycad models, and their varied insect mimics (144,146,166,167). Narrowness (“specialization”) of DT functional breadth. As indicated in this study, another distinction between the Late Triassic and Middle Jurassic is an increase in the narrowness of DT functional breadth (Figure 5). From the late Paleozoic to mid-Cretaceous ecological expansion of angiosperms, it was common for a particular feeding guild of insects to be confined to consuming only one host-plant species/morphotype within an assemblage, coupled with an increased narrowing of DT functional breadth. The reason for this trend is unclear. Because mid-Mesozoic plant assemblages included gymnosperm and later angiosperm dominated plant communities that were more species diverse than late Paleozoic communities (98,168), there was additional primary productivity created by Mesozoic plants (78). An increase in DT richness coupled with narrow DT functional breadth probably was associated with this plant-host transition through avoidance of competition from other insect herbivores (100). A possible reason for this pattern is that mouthpart class disparity approached a plateau during the Jurassic (125) (Figure 6), resulting in almost all major herbivore mouthpart classes, and by extension their feeding types, were present by the later Jurassic, before the ascendency of angiosperms (Figure 6). Summary. These five features of a diverse plant community from the ground to canopy strata, the diverse modes of insect pollination and equally diverse receptive structures of their pollinated plants, the presence of a substantial armamentarium of antiherbivore chemical and physical defenses, examples of plant–insect mimicry, and a strong tendency toward “specialization” on later Mesozoic host plants evidently were present before the ecologic expansion of angiosperms. These features provide a view of a diverse mid-Mesozoic terrestrial biota that was functionally similar to that of the Cenozoic and the present day. Appendix S10. Post 2007 updates for damage types (DTs) to the Damage Guide The updating of DTs to the current 430 DTs brings up the important issue of DT equivalency between older studies using the 2007 version 3 of the Damage Guide versus newer studies using the version 4 of the Damage Guide that includes 280 more recently discovered DTs. The additional DTs have necessitated comparisons among published plant assemblages before and after the most recent revision of the Damage Guide. Because of this update, plant assemblages using DTs described from the older version 3 of the Damage Guide (76) were categorized as to how similar they are to their closest analog in the newer version 4 of the Damage Guide. Consequently, DT001 to DT150 were compared to their closest DTs of DT151 to DT430 of version 4, whose degree of similarity was categorized as (1), (indicating high similarity); (2) (intermediate similarity), and (3) (minimal similarity). These three degrees of DT similarity were factored into our analyses. The resulting analyses were performed on these three renderings of DT categorization: the original, or raw version; the mildly lumped version for those DTs above DT150 showing a category 1 similarity or the fully lumped version for those DTs above DT150 showing category 2 or 3 similarity. This DT “standardization” was necessary to establish DT equivalencies between the older and newer DTs in studies of the plant assemblages. Appendix S11. Feeding event occurrence data that establish links between damage types and plant hosts A feeding event occurrence is a record of a particular DT presence on the surface of a plant specimen as a result of a single feeding session. This new, behavioral measure records a DT feeding bout as evidenced by plant damage resulting from arthropod consumption or from pathogen adsorption of live plant tissue. It is the finest grained level of herbivory detectible in the fossil record. Feeding event occurrences can constitute the links between DTs and their plant-host species/morphotypes; the frequency of each link in a plant assemblage provides an interaction strength that expresses the intensity of feeding between a particular DT and its plant host(s). There are recent examples of this new metric used to assess herbivory in plant assemblages (2, 5,74), although feeding event occurrence data currently have not been used to date in studies such as bipartite network webs (90). Appendix S12. The data collection, standards, and the five criteria for acceptance of plant assemblages; spreadsheet data format Fossil herbivory data were collected from online articles in Google Scholar that were searched by a combination of the key words “damage types”, “functional feeding group”, “insect herbivory”, “deep time”, and “fossil”. We obtained raw data from 134 plant assemblages (Table S1; Dataset S2), the majority of which were accessed from previously published data including three, crucial, mid-Mesozoic plant assemblages. We used five criteria for acceptance of these 134 plant assemblages. The three relevant mid Mesozoic datasets, supplied by coauthors of this study, were the late Middle Jurassic (Callovian, 165 Ma) Daohugou plant assemblage of Inner Mongolia, northeastern China (97, 169); Early Cretaceous (Barremian, 125 Ma) Dawangzhangzi plant assemblage of Liaoning, northeastern China (2); and Early Cretaceous (Albian, 103 Ma) Rose Creek plant assemblage of southeastern Nebraska, U.S.A. (3, 4). Deposition of plant (and arthropod) material were transported from within the site or from nearby sources. Other datasets originate worldwide from localities representing the past 305 million years and include 54 Late Triassic (Carnian, ca. 232 Ma) Molteno plant assemblages from Lesotho and South Africa (6, 170). Published modern datasets and fossil data (150, 151) also were part of this study. All analyses used the FFG-DT categorization herbivory data (76). Once plant fossil material was processed, 12 items of data was collected for each examined specimen and entered into a spreadsheet (Table S1). Five standards were provided for acceptance of DT data from the 134 fossil plant assemblages in our analyses. First, each plant assemblage contained minimally 300 examined plant specimens. Second, each plant species examined was represented by 20 or more specimens or alternatively a surface coverage area greater than 85% for each specimen (171), calculated using the “coverages” function in the “entropart” package (172). Third, arthropod damage was defined by DTs and not by ichnotaxa (76). Fourth, the age of the plant assemblage was determined by the narrowest available age date available and not the midpoint of a more encompassing geochronological interval, such as Callovian Age rather than the midpoint of the Middle Jurassic Epoch. Fifth, due to the uneven collection of pathogen and other poorly sampled FFGs, such DT data across plant assemblages were excluded in certain analyses. Leaf mass per area (150) was not evaluated. For each specimen, pertinent data were entered as a row of a spreadsheet. Each specimen line of the spreadsheet listed 12 columns of data, with the columns containing the following categories. Some of the 134 plant assemblages lacked data for some of the nonessential categories. 1. Plant assemblage (the plant assemblage from where the specimen originated). 2. The deposit’s age. 3. Institutional repository of the specimen. 4. Specimen catalog number. 5. Plant species/morphotype, 6. DT richness (DT types; also known as “DT diversity”). 7. DT frequency (total number of DT occurrences). 8. Feeding event occurrences. 9. Ratio of herbivorized to unherbivorized surface area. 10. Damage-type functional breadth or similar index of “specialization”. 11. Photographic log. 12. Commentary. Supporting Information References 1. L. F. Xiao, Y. K. Wu, C. C. Labandeira, L. Chen, D. Ren, Y. J. Wang, A new regime of herbivory from diverse plant–arthropod interactions from the Middle Jurassic of Northeastern China, Palaeogeogr. Palaeoclimatol. Palaeoecol. 655, 112511 (2024). 2. L. F. Xiao, C. C. Labandeira, D. Ren, Insect herbivory immediately before the eclipse of the gymnosperms: The Dawangzhangzi plant assemblage of Northeastern China. Insect Sci. 108 (2021). 3. L. F. Xiao, C. C. Labandeira, D. L. Dilcher D. Ren, Arthropod and fungal herbivory at the dawn of angiosperm diversification: The Rose Creek plant assemblage of Nebraska, USA. Cretac. Res. 131, 105088 (2022). 4. L. F. Xiao, C. C. Labandeira, D. L. Dilcher, D. Ren, Data, metrics, and methods for “Arthropod and fungal herbivory at the dawn of angiosperm diversification: The Rose Creek plant assemblage of Nebraska, USA.” Data Br. 42, 108170 (2022). 5. Q. Xu, J. Jin, C.C. Labandeira, Williamson Drive: Herbivory from a north-central Texas flora of latest Pennsylvanian age shows discrete component community structure, expansion of piercing and sucking, and plant counterdefenses. Rev. Palaeobot. Palyno. 251, 28–72 (2018). 6. C. C. Labandeira, J. M. Anderson, H. M. Anderson, “Expansion of Arthropod Herbivory in Late Triassic South Africa: The Molteno Biota, Aasvoëlberg 411 Site and Developmental Biology of a Gall” in The Late Triassic World: Earth in a Time of Transition, H.T. Lawrence Eds. (Springer Switzerland, 2018). 7. L. Azevedo-Schmidt, A. Swain, L. G. Shoemaker, E. D. Currano, Landscape-level variability and insect herbivore outbreak captured within modern forests provides a framework for interpreting the fossil record. Sci. Rep. 13, 9701 (2023). 8. T. Su et al., Resilience of plant-insect interactions in an oak lineage through Quaternary climate change. Paleobiology 41, 174–186 (2015). 9. M. A. Khan, R. A. Spicer, T. E. Spicer, S. Bera, Fossil evidence of insect folivory in the eastern Himalayan Neogene Siwalik forests. Palaeogeogr. Palaeoclimatol. Palaeoecol. 410, 264–277 (2014). 10. B. Adroit et al., Plant-insect interactions patterns in three European paleoforests of the late-Neogene– early-Quaternary. PeerJ 6, e5075 (2018). 11. M. Hazra et al., Pliocene Albizia (Fabaceae) from Jharkhand, eastern India: reappraisal of its biogeography during the Cenozoic in Southeast Asia. Palaeoworld 31, 153–168 (2021). 12. J. M. Robledo, E. R. Pinheiro, S. C. Gnaedinger, T. Wappler, Plant-insect interactions on dicots and ferns from the Miocene of Argentina. Palaios 33, 338–352 (2018). 13. T. Wappler, F. Grímsson, Before the 'Big Chill': Patterns of plant-insect associations from the Neogene of Iceland. Global Planet. Change 142, 73–86 (2016). 14. S. H., Zhang et al., Plant–insect associations from the upper Miocene of Lincang, Yunnan, China. Rev. Palaeobot. Palyno. 259, 55–62 (2018). 15. G. Zidianakis, J. Kovar-Eder, A. Zelilidis, G. Iliopoulos, Evidence of plant–arthropod interaction in the fossil assemblage from Pitsidia (Messara Basin, Crete, Greece; Upper Miocene). Neues Jahrb. Geol. Paläont.-Abh. 295, 101–115 (2020). 16. T. Reichgelt et al., Diverse and abundant insect herbivory on Miocene Nothofagaceae of the Dunedin Volcano, Otago, New Zealand. Palaeobio. Palaeoenv. 96, 265–279 (2016). 17. Q. Li et al., High frequency of arthropod herbivore damage in the Miocene Huaitoutala flora from the Qaidam Basin, northern Tibetan Plateau. Rev. Palaeobot. Palyno. 297, 104569 (2022). 18. M. A. Khan, S. Bera, R. A. Spicer, T. E. Spicer, Plant–arthropod associations from the Siwalik forests (middle Miocene) of Darjeeling sub-Himalaya, India. Palaeogeogr. Palaeoclimatol. Palaeoecol. 438, 191–202 (2015). 19. J. L. Dong, The Miocene plant fossils, paleoatmospheric CO2 and plant-insect interactions from Fujian province. (School of Earth Sciences, Lanzhou University, PhD thesis, 2019). 20. S. X. Yin, X. T. Wu, Z. X. Wang, G. L. Shi. “A preliminary study on the Plant–Insect ecological relationship of the Miocene Tropical Forest in Zhangpu Biota, Fujian.” The 2021 Academic Annual Meeting of the Paleontology Branch of the Chinese Palaeontological Society and the Jiangsu Paleontological Society. Changchun: Jilin. (2021). 21. F. J. Ma, et al., Plant–insect interactions from the Miocene (Burdigalian–Langhian) of Jiangxi, China. Rev. Palaeobot. Palyno. 275, 104176 (2020). 22. M. R. Carvalho, et al., Extinction at the end-Cretaceous and the origin of modern Neotropical rainforests. Science 372, 63–68 (2021). 23. B. Adroit, V. Teodoridis, T. H., Güner, T. Denk, Patterns of insect damage types reflect complex environmental signal in Miocene forest biomes of Central Europe and the Mediterranean. Global Planet. Change 199, 103451 (2021). 24. A. L. Möller, U. Kaulfuss, D. E. Lee, T. Wappler, High richness of insect herbivory from the early Miocene Hindon Maar crater, Otago, New Zealand. PeerJ 5, e2985 (2017). 25. S. Knor, Z. Kvaček, T. Wappler, J. Prokop, Diversity, taphonomy and palaeoecology of plant– arthropod interactions in the lower Miocene (Burdigalian) in the Most Basin in north-western Bohemia (Czech Republic). Rev. Palaeobot. Palyno. 219, 52–70 (2012). 26. E. D. Currano, B. F Jacobs, Bug-bitten leaves from the early Miocene of Ethiopia elucidate the impacts of plant nutrient concentrations and climate on insect herbivore communities. Global Planet. Change 207, 103655 (2021). 27. T. Wappler, Insect herbivory close to the Oligocene-Miocene transition – A quantitative analysis. Palaeogeogr. Palaeoclimatol. Palaeoecol. 292, 540–550 (2010). 28. S. Gunkel, T. Wappler, Plant-insect interactions in the upper Oligocene of Enspel (Westerwald, Germany), including an extended mathematical framework for rarefaction. Palaeobio. Palaeoenv. 95, 55–75 (2015). 29. E. D. Currano, B. F. Jacobs, A. D. Pan, N. J. Tabor, Inferring ecological disturbance in the fossil record: A case study from the late Oligocene of Ethiopia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 309, 242–252 (2011). 30. R. Moreno-Domínguez, Primeras interacciones planta-insecto del Oligoceno de la Península Ibérica. Revista de la Sociedad Geológica de España 31, 19–28 (2018). 31. W. Deng et al., Sharp changes in plant diversity and plant-herbivore interactions during the Eocene– Oligocene transition on the southeastern Qinghai-Tibetan Plateau. Global Planet. Change 194, 103293 (2020). 32. T. Wappler, T. Denk, Herbivory in early Tertiary Arctic forests. Palaeogeogr. Palaeoclimatol. Palaeoecol. 310, 283–295 (2011). 33. L. Kunzmann et al., A Paleogene leaf flora (Profen, Sachsen-Anhalt, Germany) and its potentials for palaeoecological and palaeoclimate reconstructions. Flora 254, 71–87 (2019). 34. C. Müller, T. Wappler, L. Kunzmann, Insect herbivory patterns in late Eocene coastal lowland riparian associations from central Germany. Palaeogeogr. Palaeoclimatol. Palaeoecol. 491, 170–184 (2018). 35. T. Wappler et al., Testing for the effects and consequences of mid Paleogene climate change on insect herbivory. PLoS ONE 7, e40744 (2012). 36. P. Wilf, C. C. Labandeira, Response of plant-insect associations to Paleocene-Eocene warming. Science 284, 2153–2156 (1999). 37. E. D. Currano et al., Endemism in Wyoming plant and insect herbivore communities during the early Eocene hothouse. Paleobiology 45, 421–439 (2019). 38. C. C. Labandeira, Paleobiology of middle Eocene plant-insect associations from the Pacific Northwest: A preliminary report. Rocky Mt. Geol. 37, 31–59 (2002). 39. P. Wilf, C. C. Labandeira, K. R. Johnson, N. R. Cúneo, Richness of plant–insect associations in Eocene Patagonia: a legacy for South American biodiversity. Proc. Natl. Acad. Sci. U.S.A. 102, 8944–8948 (2005). 40. E. D. Currano, C. C. Labandeira, P. Wilf, Fossil insect folivory tracks paleotemperature for six million years. Ecol. Monogr. 80, 547–567 (2010). 41. P. Wilf, C. C. Labandeira, K. R. Johnson, B. Ellis, Decoupled plant and insect diversity after the end- Cretaceous extinction. Science 313, 1112–1115 (2006). 42. E. D. Currano, et al., Sharply increased insect herbivory during the Paleocene–Eocene Thermal Maximum. Proc. Natl. Acad. Sci. U.S.A. 105, 1960–1964 (2008). 43. L. E. A. Schmidt et al., Plant and insect herbivore community variation across the Paleocene–Eocene boundary in the Hanna Basin, southeastern Wyoming. PeerJ 7, e7798 (2019). 44. S. L. Wing et al., Late Paleocene fossils from the Cerrejón Formation, Colombia, are the earliest record of Neotropical rainforest. Proc. Natl. Acad. Sci. U.S.A. 106, 18627–18632 (2009). 45. T. Wappler et al., No post-Cretaceous ecosystem depression in European forests? Rich insect-feeding damage on diverse middle Palaeocene plants, Menat, France. Proc. R. Soc. B Biol. Sci. 276, 4271– 4277 (2009). 46. M. P. Donovan, et al., Rapid recovery of Patagonian plant–insect associations after the end- Cretaceous extinction. Nat. Ecol. Evol. 1, 1–5 (2016). 47. M. P. Donovan et al., Diverse plant–insect associations from the latest Cretaceous and early Paleocene of Patagonia, Argentina. Ameghiniana 55, 303–338 (2018). 48. M. P. Donovan, et al., Novel insect leaf-mining after the end-Cretaceous extinction and the demise of Cretaceous leaf miners, Great Plains, USA. PLoS ONE 9, 1–35 (2014). 49. C. C. Labandeira, K. R., Johnson, P. Wilf, Impact of the terminal Cretaceous event on plant–insect associations. Proc. Natl. Acad. Sci. U.S.A. 99, 2061–2066 (2002). 50. M. P. Donovan, et al., Persistent biotic interactions of a Gondwanan conifer from Cretaceous Patagonia to modern Malesia. Commun. Biol. 3, 708 (2020). 51. C. C. Labandeira, E. Kustatscher, T. Wappler, Floral assemblages and patterns of insect herbivory during the Permian to Triassic of Northeastern Italy. PLoS ONE 11, e0165205 (2016a). 52. S. A. Maccracken, J. C. Sohn, I. M. Miller, C. C. Labandeira, A new Late Cretaceous leaf mine Leucopteropsa spiralae gen. et sp. nov. (Lepidoptera: Lyonetiidae) represents the first confirmed fossil evidence of the Cemiostominae. J. Syst. Palaeontol. 19, 131–144 (2021). 53. S. Maccracken et al., Insect herbivory on Catula gettyi gen. et sp. nov. (Lauraceae) from the Kaiparowits Formation (Late Cretaceous, Utah, USA). PLoS ONE 17, e0261397 (2022). 54. A. A. Santos, et al., Plant-insect interactions from the mid-Cretaceous at Puy-Puy (Aquitaine Basin, western France) indicates preferential herbivory for angiosperms amid a forest of ferns, gymnosperms, and angiosperms. Bot. Lett. 169, 568–587 (2022). 55. A. A. Santos et al., A Robinson Crusoe story in the fossil record: Plant-insect interactions from a Middle Jurassic ephemeral volcanic island (Eastern Spain). Palaeogeogr. Palaeoclimatol. Palaeoecol. 583, 110655 (2021). 56. Y. Y. Xu et al., Plant-insect interactions across the Triassic–Jurassic boundary in the Sichuan Basin, South China. Front. Ecol. Evol. 11, 1338865 (2024). 57. Q. Ding, C. C. Labandeira, Q. Meng, D. Ren, Insect herbivory, plant-host specialization and tissue partitioning on mid-Mesozoic broadleaved conifers of Northeastern China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 440, 259–273 (2015). 58. Y. X. Chen, The Late Triassic Marattiaceae and Dipteridaceae and their palaeoecology in Yipinglang, Yunnan University, master’s thesis. (2016). 59. J. J. Shen, The Late Triassic Podozamites and Ferganiella from Lufeng, Yunnan University, college of palaeoecology. Master's thesis. (2017). 60. H. Y. Liu, et al., A latitudinal gradient of plant–insect interactions during the late Permian in terrestrial ecosystems? New evidence from Southwest China. Global Planet. Change 192, 103248 (2020). 61. R. Prevec, et al., Portrait of a Gondwanan ecosystem: a new late Permian fossil locality from KwaZulu-Natal, South Africa. Rev. Palaeobot. Palyno. 156, 454–493 (2009). 62. J. M. Anderson, H. M. Anderson. Palaeoflora for Southern Africa: Prodromus of South African megafloras. Devonian to Lower Cretaceous (Balkema, Rotterdam, 1985). 63. S. A. Maccracken, C. C. Labandeira, The Middle Permian South Ash Pasture assemblage of north- central Texas: Coniferophyte and gigantopterid herbivory and longer-term herbivory trends. Int. J. Plant Sci. 181, 342–362 (2020). 64. B. Cariglino, Patterns of insect-mediated damage in a Permian Glossopteris flora from Patagonia (Argentina). Palaeogeogr. Palaeoclimatol. Palaeoecol. 507, 39–51 (2018). 65. S. R. Schachat et al., Plant-insect interactions from early Permian (Kungurian) Colwell Creek Pond, north-central Texas: the early spread of herbivory in riparian environments. Int. J. Plant Sci. 175, 855–890 (2014). 66. K. Adami-Rodrigues, R. Iannuzzi, I. D. Pinto, “Permian plant-insect interactions from a Gondwana flora of southern Brazil.” In Trace fossils in evolutionary palaeoecology: Proceedings of Session 18 (Trace Fossils) of the First International Palaeontological Congress Sydney, Australia, 106–125 (2004). 67. E. R. Pinheiro, R. Iannuzzi, G. P. Tybusch, Specificity of leaf damage in the Permian “Glossopteris Flora”: a quantitative approach. Rev. Palaeobot. Palyno. 174, 113–121 (2012). 68. S. R. Schachat, C. C. Labandeira, D. S. Chaney, Insect herbivory from early Permian Mitchell Creek Flats of north–central Texas: opportunism in a balanced component community. Palaeogeogr. Palaeoclimatol. Palaeoecol. 440, 830–847 (2015). 69. A. L. Beck, C. C. Labandeira, Early Permian insect folivory on a gigantopterid-dominated riparian flora from north-central Texas. Palaeogeogr. Palaeoclimatol. Palaeoecol. 142, 139–173 (1998). 70. Z. Feng et al., Plant–insect interactions in the early Permian Wuda Tuff Flora, North China. Rev. Palaeobot. Palyno. 294, 104269 (2021). 71. Banerjee, M., & Bera, S. (1998). Record of zoocecidia on leaves of Glossopteris browniana Brongn. from Mohuda Basin, Upper Permian, Indian Lower Gondwana. Indian Biologist, 30, 58–61. 72. J. Gallego, R. Cuneo, I. Escapa, Plant–arthropod interactions in gymnosperm leaves from the Early Permian of Patagonia, Argentina. Geobios 47, 101–110 (2014). 73. C. C. Labandeira, E. G. Allen, Minimal insect herbivory for the Lower Permian Coprolite Bone Bed site of north-central Texas, USA, and comparison to other Late Paleozoic floras. Palaeogeogr. Palaeoclimatol. Palaeoecol. 247, 197–219 (2007). 74. T. B. Dos Santos, C. C. Labandeira, E. R. S. Pinheiro, R. Iannuzzi, Plant interactions with arthropods and pathogens at Sanzenbacher Ranch, early Permian of Texas, and implications for herbivory evolution in Southwestern Euramerica. Front. Ecol. Evol. 12, 1368174 (2024). 75. M. P. Donovan, S. G. Lucas, Insect herbivory on the Late Pennsylvanian Kinney Brick Quarry flora, New Mexico, USA. New Mexico Museum of Natural History and Science Bulletin, 84, 193–207 (2021). 76. C. C. Labandeira, P. Wilf, K. R. Johnson, F. Marsh, Guide to insect (and other) damage types on compressed plant fossils (version 3.01). (Smithsonian Institution, National Museum of Natural History, Department of Paleobiology, Washington, DC., 2007). 77. F. Berendse, M. Scheffer, The angiosperm radiation revisited, an ecological explanation for Darwin’s ‘abominable mystery.’ Ecol. Lett. 12, 865–872 (2009). 78. E. Biffin, et al., Leaf evolution in Southern Hemisphere conifers tracks the angiosperm ecological radiation. Proc. R. Soc. B Biol. Sci. 279, 341–348 (2012). 79. W. J. Bond, The tortoise and the hare: ecology of angiosperm dominance and gymnosperm persistence. Biol. J. Linn. Soc. 36, 227–249 (1989). 80. I. Porth, B. Hamberger, R. White, K. Ritland, Defense mechanisms against herbivory in Picea: sequence evolution and expression regulation of gene family members in the phenylpropanoid pathway. BMC Genomics 12, 608 (2011). 81. M. Wink, “Evolution of the angiosperms and co-evolution of secondary metabolites, especially of alkaloids” in: Co-Evolution of Secondary Metabolites. J-M. Mérillon, K. G. Ramawat Eds. (Springer Switzerland, 2020), pp. 1–24. 82. F. L. Condamine, M. E. Clapham, G. J. Kergoat, Global patterns of insect diversification: towards a reconciliation of fossil and molecular evidence? Sci. Rep. 6, 19208 (2016). 83. E. L. Zvereva, M. V. Kozlov, Consequences of simultaneous elevation of carbon dioxide and temperature for plant–herbivore interactions: a metaanalysis. Global Change Biol. 12, 27–41 (2006). 84. L. Augusto, T. J. Davies, S. Delzon, A. De Schrijver, The enigma of the rise of angiosperms: can we untie the knot? Ecol. Lett. 17, 1326–1338 (2014). 85. M. J. Benton, P. Wilf, H. Sauquet, The Angiosperm Terrestrial Revolution and the origins of modern biodiversity. New Phytol. 233, 2017–2035. 86. C. C. Labandeira, The four phases of plant–arthropod associations in deep time. Geologica Acta, 4, 409–438 (2006a). 87. C. C. Labandeira, T. Wappler, Arthropod and pathogen damage on fossil and modern plants: exploring the origins and evolution of herbivory on land. Annu. Rev. Entomol. 68, 341–361 (2023). 88. C. C. Labandeira, Deep–time patterns of tissue consumption by terrestrial arthropod herbivores. Naturwissenschaften 100, 355–364 (2013). 89. L. A. Buatois, et al., The invasion of the land in deep time: Integrating Paleozoic records of paleobiology, ichnology, sedimentology, and geomorphology. Integr. Comp. Biol. 62, 297–331 (2022). 90. A. Swain, S. A. Maccracken, W. F. Fagan, C. C. Labandeira, Understanding the ecology of plant– insect herbivore interactions in the fossil record through bipartite networks. Paleobiology 48, 239– 260 (2022). 91. C. C. Labandeira, The origin of herbivory on land: initial patterns of plant tissue consumption by arthropods. Insect Sci. 14, 259–275 (2007). 92. C. C. Labandeira, S. L. Tremblay, K. E. Bartowski, L. VanAller Hernick, Middle Devonian liverwort herbivory and antiherbivore defence. New Phytol. 202, 247–258 (2014). 93. R. J. Knecht et al., Early Pennsylvanian Lagerstätte reveals a diverse ecosystem on a subhumid, alluvial fan. Nature Commun. 15, 7876. 94. C. C. Labandeira, Early history of arthropod and vascular plant associations. Annu. Rev. Earth Pl. Sc. 26, 329–377 (1998). 95. C. C. Labandeira, E. D. Currano, The fossil record of plant–insect dynamics. Annu. Rev. Earth Pl. Sc. 41, 287–311 (2013). 96. X. Lin, et al., Exploiting nondietary resources in deep time: Patterns of oviposition on mid-Mesozoic plants from Northeastern China. Int. J. Plant Sci. 180, 411–457 (2019). 97. L. F. Xiao et al., Middle Jurassic insect mines on gymnosperms provide missing links to early mining evolution. New Phytol. 242, 2391–2393 (2024). 98. M. M. Turcotte, T. J. Davies, C. J. Thomsen, M. T. Johnson, Macroecological and macroevolutionary patterns of leaf herbivory across vascular plants. Proc. R. Soc. B Biol. Sci. 281, 20140555 (2014). 99. D. Peris, F. Condamine, The angiosperm radiation played a dual role in the diversification of insects and insect pollinators. Nat. Commun. 15, 1–11 (2024). 100. C. C. Labandeira, “Why did terrestrial insect diversity not increase during the angiosperm radiation? Mid–Mesozoic, plant–associated insect lineages harbor clues” in: Evolutionary Biology: Genome Evolution, Speciation, Coevolution and Origin of Life, P. Pontarotti Ed. (Springer Switzerland, 2014), pp. 261–299. 101. J. Rota, et al., The unresolved phylogenomic tree of butterflies and moths (Lepidoptera): assessing the potential causes and consequences. Syst. Entomol. 47, 531–550 (2022). 102. N. Song, et al., 300 million years of diversification: elucidating the patterns of orthopteran evolution based on comprehensive taxon and gene sampling. Cladistics 31, 621–651 (2015). 103. J. Szwedo, Paradise Lost? – Cretaceous and Palaeogene diversification of planthoppers and leafhoppers. Bull. Insectology 61, 111–112 (2008). 104. D. Ahrens, J. Schwarzer, A. P. Vogler, The evolution of scarab beetles tracks the sequential rise of angiosperms and mammals. Proc. R. Soc. B Biol. Sci. 281, 20141470 (2014). 105. J. Gómez-Zurita, T. Hunt, F. Kopliku, A. P. Volger, Recalibrated tree of leaf beetles (Chrysomelidae) indicates independent diversification of angiosperms and their insect herbivores. PLoS ONE 2, e360 (2007). 106. S. Winter, A. L. L. Friedman, J. J. Astrin, B. Gottsberger, H. Letsch, Timing and host plant associations in the evolution of the weevil tribe Apionini (Apioninae, Brentidae, Curculionoidea, Coleoptera) indicate an ancient co-diversification pattern of beetles and flowering plants. Mol. Phylogenet. Evol. 107,179–190 (2017). 107. I. S. Winkler, C. Mitter, S. J. Scheffer, Repeated climate-linked host shifts have promoted diversification in a temperate clade of leaf-mining flies. Proc. Natl. Acad. Sci. U.S.A. 106, 18103– 18108 (2009). 108. C. Doorenweerd, E. J. van Nieukerken, R. J. B. Hoare, Phylogeny, classification and divergence times of pygmy leaf-mining moths (Lepidoptera: Nepticulidae): the earliest lepidopteran radiation on Angiosperms? Syst. Entomol. 42, 267–287 (2017). 109. G., Fagua, et al., Convergent herbivory on conifers by Choristoneura moths after boreal forest formation. Mol. Phylogenet. Evol. 123, 35–43 (2018). 110. N. Wahlberg, C. W. Wheat, C. Peña, Timing and patterns in the taxonomic diversification of Lepidoptera (butterflies and moths). PLoS ONE 8, e80875 (2013). 111. C. S. Moreau, et al., Phylogeny of the ants: Diversification in the age of angiosperms. Science 312, 101–104 (2006). 112. S. Cardinal, B. M Danforth, Bees diversified in the age of eudicots. Proc. R. Soc. B Biol. Sci. 280, 20122686 (2013). 113. E. Romero-Lebrón et al., Endophytic insect oviposition traces in deep time. Palaeogeogr. Palaeoclimatol. Palaeoecol. 590, 110855 (2022). 114. J. F. Tooker, D. Giron, The evolution of endophagy in herbivorous insects. Front. Plant Sci. 11, 581816 (2020). 115. Y. Asar, S. Y. Ho, H. Sauquet, Early diversifications of angiosperms and their insect pollinators: were they unlinked? Trends Plant Sci. 27, 858–869 (2022). 116. C. Peña–Kairath, et al., Insect pollination in deep time. Trends Ecol. Evol. 38, 749–759 (2023). 117. D. Ren et al., A probable pollination mode before angiosperms: Eurasian, long–proboscid scorpionflies. Science 326, 840–847 (2009). 118. D. Peris et al., The case of Darwinylus marcosi (Insecta: Coleoptera: Oedemeridae), A Cretaceous shift from a gymnosperm to an angiosperm pollinator mutualism. Commun. Integr. Biol. 10, 897–904 (2017). 119. D., Peris et al., Generalist Pollen–Feeding Beetles during the Mid–Cretaceous. iScience 23, 100913 (2020). 120. M. G. Mángano, L. A. Buatois, (2016). The Trace–Fossil Record of Major Evolutionary Events: Volume 2: Mesozoic and Cenozoic (Vol. 40). (Springer Switzerland, 2016) 121. C. C. Labandeira, L. F. Li, “The history of insect parasitism and the Mid-Mesozoic Parasitoid Revolution” in: The Evolution and Fossil Record of Parasitism—Identification and Macroevolution of Parasites. K. De Baets, J. Huntley Eds. Topics in Geobiology 49, 377–533 (2021). 122. B. Wang, C. Xu, E. A. Jarzembowski, Ecological radiations of insects in the Mesozoic. Trends Ecol. Evol. 37, 529–540 (2022). 123. E. A. Bernays, Evolution of feeding behavior in insect herbivores. Bioscience 48, 35–44 (1998). 124. C. C. Labandeira, Insect mouthparts: ascertaining the paleobiology of insect feeding strategies. Annu. Rev. Ecol. Syst. 28, 153–193 (1997). 125. C.C. Labandeira, “The Fossil Record of Insect Mouthparts: Innovation, Functional Convergence, and Associations with Other Organisms” in Insect Mouthparts Zoological Monographs 5, H.W. Krenn Ed. (Springer Switzerland, 2019), pp. 567–671. 126. M. L. Forister, et al., The global distribution of diet breadth in insect herbivores. Proc. Natl. Acad. Sci. U.S.A. 112, 442–447 (2015). 127. C. Du, J. Chen, L. Jiang, G. Qiao, High correlation of species diversity patterns between specialist herbivorous insects and their specific hosts. J. Biogeogr. 47, 1232–1245 (2020). 128. M. V. Kozlov, V. Lanta, V. Zverev, E. L. Zvereva, Global patterns in background losses of woody plant foliage to insects. Global Ecol. Biogeogr. 24, 1126–1135 (2015). 129. J. S. Bale et al., Herbivory in global climate change research: direct effects of rising temperature on insect herbivores. Global Change Biol. 8, 1–16 (2002). 130. E. L. Zvereva, V. Zverev, V. A. Usoltsev, M. V. Kozlov, Latitudinal pattern in community‐wide herbivory does not match the pattern in herbivory averaged across common plant species. J. Ecol. 108, 2511–2520 (2020). 131. H. Tang et al., Global latitudinal patterns in leaf herbivory are related to variation in climate, rather than phytochemicals or mycorrhizal types. Natl. Sci. Rev. 10, nwad236 (2023). 132. A. Baselga, Separating the two components of abundance-based dissimilarity: balanced changes in abundance vs. abundance gradients. Methods Ecol. Evol. 4, 552–557 (2013). 133. A. Baselga, Partitioning abundance-based multiple-site dissimilarity into components: Balanced variation in abundance and abundance gradients. Methods Ecol. Evol. 8, 799–808 (2017). 134. H., Jiang, et al., Mesozoic evolution of cicadas and their origins of vocalization and root feeding. Nat. Commun. 15, 376 (2024). 135. I. M. Vea, D. A. Grimaldi, Putting scales into evolutionary time: the divergence of major scale insect lineages (Hemiptera) predates the radiation of modern angiosperm hosts. Sci. Rep. 6, 23487 (2016). 136. N. Song, A. P. Liang, A preliminary molecular phylogeny of planthoppers (Hemiptera: Fulgoroidea) based on nuclear and mitochondrial DNA sequences. PLoS ONE 8, e58400 (2013). 137. B., Wang, Zhang H., E. A. Jarzembowski, Early Cretaceous angiosperms and beetle evolution. Front. Plant Sci. 4, 360 (2013). 138. D. D. McKenna, et al., The evolution and genomic basis of beetle diversity. Proc. Natl. Acad. Sci. U.S.A. 116, 24729–24737 (2019). 139. W., Zhang, et al., New fossil Lepidoptera (Insecta: Amphiesmenoptera) from the Middle Jurassic Jiulongshan Formation of Northeastern China. PLoS ONE 8, e79500 (2013). 140. T. J. B. van Eldijk, et al., A Triassic–Jurassic window into the evolution of Lepidoptera. Sci. Adv. 4, e1701568 (2018). 141. T. Nyman, et al., The early wasp plucks the flower: disparate extant diversity of sawfly superfamilies (Hymenoptera: ‘Symphyta’) may reflect asynchronous switching to angiosperm hosts. Biol. J. Linn. Soc. 128, 1–19 (2019). 142. M. Wang et al., Mirolydidae, a new family of Jurassic pamphilioid sawfly (Hymenoptera) highlighting mosaic evolution of lower Hymenoptera. Sci. Rep. 7, 43944 (2017). 143. D. Lai et al., The associated evolution of raptorial foreleg and mantispid diversification during 200 million years. Natl. Sci. Rev. 10, nwad278 (2023). 144. C. C. Labandeira et al., evolutionary convergence of mid-Mesozoic lacewings and Cenozoic butterflies. Proc. R. Soc. B Biol. Sci. 283, 20152893 (2016). 145. D. Ren et al., Rhythms of insect evolution: evidence from the Jurassic and Cretaceous in northern China. (John Wiley & Sons Ltd., New York, 2019). 146. H. Fang et al., Lichen mimesis in mid-Mesozoic lacewings. eLife 9, e59007 (2020). 147. S. R. Schachat, S. A. Maccracken, C. C. Labandeira, Sampling fossil floras for the study of insect herbivory: how many leaves is enough? Mitteilungen aus dem Museum für Naturkunde in Berlin: The Fossil Record 23, 15–32 (2020). 148. J. Albrecht, T. Wappler, S. A. Fritz, M. Schleuning, Fossil leaves reveal drivers of herbivore functional diversity during the Cenozoic. Proc. Natl. Acad. Sci. U.S.A. 120, e2300514120 (2023). 149. M. R., Carvalho, et al., Insect leaf–chewing damage tracks herbivore richness in modern and ancient forests. PLoS ONE 9, e94950 (2014). 150. L. Azevedo–Schmidt, E. K. Meineke, E. D. Currano, Insect herbivory within modern forests is greater than fossil localities. Proc. Natl. Acad. Sci. U.S.A. 119, e2202852119 (2022). 151. E. D. Currano, L. E. Azevedo–Schmidt, S. A. Maccracken, A. Swain, Scars on fossil leaves: An exploration of ecological patterns in plant–insect herbivore associations during the Age of Angiosperms. Palaeogeogr. Palaeoclimatol. Palaeoecol. 582, 110636 (2021). 152. Y. Na et al., A brief introduction to the Middle Jurassic Daohugou Flora from Inner Mongolia, China. Rev. Palaeobot. Palyno. 247, 53–67 (2017). 153. P. Wilf et al., Insect herbivory, plant defense, and early Cenozoic climate change. Proc. Natl. Acad. Sci. U.S.A. 98, 6221–6226 (2001). 154. D. L. Royer et al., Fossil leaf economics quantified: calibration, Eocene case study, and implications. Paleobiology 33, 574–589 (2007). 155. W. K. Soh et al., Palaeo leaf economics reveal a shift in ecosystem function associated with the end- Triassic mass extinction event. Nat. Plants 3, 17104.l (2017). 156. L. L. Veromann–Jürgenson et al., Predictability of leaf morphological traits for paleoecological reconstruction: the case of leaf cuticle and leaf dry mass per area. Int. J. Plant Sci. 181, 129–141 (2020). 157. D. Y. Huang, The Daohugou Biota. (Shanghai Science and Technology Press, 2016). 158. K. G. Sun, J. Z Cui, S. Wang, Fossil Gymnosperms in China. Volume 1. (Beijing: Higher Education Press, 2016b). 159. C. Pott, B. Jiang, Plant remains from the middle–late Jurassic Daohugou site of the Yanliao biota in Inner Mongolia. Acta Palaeobotanica 57, 185–222 (2017). 160. P.R. Crane, E.M. Friis, K.R. Pedersen, The origin and early diversification of angiosperms. Nature 374, 27–33 (1995). 161. T. Salles, L. Husson, M. Lorcery, B. Hadler-Boggiani, Landscape dynamics and the Phanerozoic diversification of the biosphere. Nature 624, 115–121 (2023). 162. K. R. Hale, F. S. Valdovinos, N. D. Martinez, Mutualism increases diversity, stability, and function of multiplex networks that integrate pollinators into food webs. Nat. Commun. 11, 2182 (2020). 163. H. Honda, “Ginkgo and insects” in Ginkgo Biloba A Global Treasure: From Biology to Medicine, T. Hori et al., Eds. (Springer Tokyo, 1997), pp. 243–250. 164. C. Castillo-Guevara, V. Rico-Gray, The role of macrozamin and cycasin in cycads (Cycadales) as antiherbivore defenses. J. Torrey Bot. Soc. 130, 206–217 (2003). 165. R. Garrouste et al., Insect mimicry of plants dates back to the Permian. Nat. Commun. 7, 13735 (2016). 166. Y. Wang et al., Jurassic mimicry between a hangingfly and a ginkgo from China. Proc. Natl. Acad. Sci. U.S.A. 109, 20514–20519 (2012). 167. H. Yang et al., Early specializations for mimicry and defense in a Jurassic stick insect. Natl. Sci. Rev. 8, nwaa056 (2021). 168. S. L. Wing, W. A. DiMichele, Conflict between local and global changes in plant diversity through geological time. Palaios 551–564 (1995). 169. L. F. Xiao et al., Early Cretaceous mealybug herbivory on a laurel highlights the deep-time history of angiosperm–scale insect associations. New Phytol. 232, 1414–1423 (2021). 170. J. M. Anderson, H. M. Anderson, C. J. Cleal, Brief history of the gymnosperms: Classification, biodiversity, phytogeography and ecology. Strelitzia 20, 1–280 (2007). 171. A. Chao, L. Jost, Coverage‐based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology 93, 2533–2547 (2012). 172. E. Marcon, B. Herault, M. E. Marcon, Package ‘entropart’ (2023).