打开冰箱看一眼：一只虾，一枚蛤蜊，一条鱼。它们身上几乎没有一处对得上——虾把硬壳套在外面、肉长在里面；蛤蜊把自己关进两片壳当中；鱼的骨头长在身体正中间。三套完全不同的搭法。
这三套，连同今天几乎所有动物的基本样式，是在同一段时间里定下来的：距今大约五亿四千万到五亿两千万年。在那之前的海里，你一套也找不到。
一堆事挤在这么短一段里发生，通常只有两种可能：要么那段时间真的出了什么事，要么是我们看见它的方式在那一刻变了。这两种在这里都占一部分，麻烦的是分清各占多少。
化石不是死了就有。软的东西烂得快，能留下来的多半是硬的：壳、刺、牙、骨片。而恰好在寒武纪最开始那一段，海里的硬部件从几乎没有变成到处都是。这句话的含义很直白——不一定是生物突然出现，至少有一部分是「能被记下来的生物」突然多了。灯是那时候开的。
那之前也不是空的。埃迪卡拉纪（大约六亿三千五百万到五亿三千九百万年前）的海底躺着一批东西：扁的、软的，有的像叶片，有的像毯子摊开，很多至今连该归到哪一类都定不下来。
那怎么知道不是全靠记录变化撑起来的？靠一类不挑硬壳的证据：遗迹化石，也就是动物在泥沙里留下的爬痕和洞。它记的是行为，身体软也照样能留下。而它在寒武纪开头那一段的变化很硬气：从埃迪卡拉纪那种贴着海底表面横划的简单拖痕，变成往下钻、会分叉、在沉积物里三维探来探去的洞。这类分叉探洞密集出现，正是国际上用来划寒武纪起点的标志之一。
这一钻还顺手改了海底本身：在那之前海底被细菌铺成的席子盖着，底下板结、缺氧；一旦有东西开始往下翻，沉积物变松、氧气渗得进去，一整批新活法才有地方站——这件事叫寒武纪底质革命。钻洞既是结果，又是下一轮的原因。
第三条线索方向相反：分子钟，也就是拿现存动物之间的基因差异倒推它们分家的年代。它给出的年代普遍比化石早得多——动物的共同祖先常落在七亿年上下，两侧对称的那些大分支多半在埃迪卡拉纪就分开了。误差很宽：2015 年一项专门检查它的研究说得直白，化石标定不确定、分子变化的速率也不恒定，精度有个绕不过去的上限。
三条拼起来，图像就清楚了：家谱早就分开了；爆发的不是「分家」，是变大、变硬、长出眼睛、开始互相吃。而且这段爆发本身也不短——按现在的估计它持续了一两千万年，等到化石变得常见（约五亿两千万年前），最激烈的那一段其实已经过去了。
EN Fossils need hard parts. Soft bodies rot; shells, spines, teeth and plates survive. Right at the start of the Cambrian, hard parts in the sea went from almost absent to everywhere — so at least part of the "sudden appearance" is that the record switched its light on. What came before was not empty: the Ediacaran seafloor (roughly 635 to 539 million years ago) held flat, soft things, some leaf-like, some spread out like blankets, many of which still cannot be confidently placed in any group. How do we know the pattern is not entirely a sampling artefact? From a line of evidence that does not need shells: trace fossils, the trails and burrows animals leave in sediment. They record behaviour, and they changed sharply — from simple horizontal trails scratched along the surface to branching, three-dimensional burrows probing down into the sediment. Dense branching burrows of that kind are one of the markers used to define the base of the Cambrian. That digging also changed the seafloor itself: before it was sealed under bacterial mats, packed and oxygen-poor; once things started turning it over, sediment loosened and oxygen got in, opening room for new ways of living — the Cambrian substrate revolution. Burrowing is both consequence and next cause. A third line points the other way: molecular clocks, which back-date lineage splits from genetic differences and generally land far earlier than fossils, with the common ancestor of animals often around 700 million years ago. That method has wide error bars, and a 2015 study checking it concluded bluntly that uncertain fossil calibrations and non-constant rates put a hard ceiling on the precision achievable. Together: the family tree split long before; what exploded was getting big, getting hard, growing eyes, and starting to eat each other. And the explosion itself lasted something like ten to twenty million years — by the time fossils become common, around 521 million years ago, the fiercest part was already over.
最常听到的解释是氧气：大的、能游、要追着别人吃的身体，耗氧比躺着不动的高得多，所以得等海水里氧够了才办得到。讲得通，但证据没有一边倒。有地球化学指标支持那前后氧在升，也有研究显示那段时间的氧不是一级台阶，而是一波一波涨了又落；2019 年一项把化石与环境记录并排放的工作干脆认为，埃迪卡拉和寒武纪没那么截然两分。稳妥的读法是：氧更像抬高门槛的背景条件，不是按一下就响的扳机。这个至今还在争。
第二个解释有意思得多：眼睛。三叶虫的眼睛是已知最早的一批高分辨率眼睛，镜片直接由方解石晶体构成，能在化石上一颗颗数出来。而海里一旦有东西能看清几十厘米外有什么，游戏就变了：躲、埋进泥里、长壳长刺、跑得更快，在同一天全都变得有用；防御一升级，又反过来逼着捕食那边加码。「红皇后」那期讲的军备竞赛，在这里第一次大规模开跑。
会不会是「新的遗传工具到货了」？这条基本被排除。比较各类动物的基因组会发现，那套控制身体分节、决定哪一段长什么的调控基因，在寒武纪之前就已经在了。工具箱是旧的，变的是接线方式和外头的机会。
所以「寒武纪大爆发的原因是什么」这个问法本身就有点歪：它要一个扳机，而这里更像一个圈转了起来。各假说的权重，学界至今没谈拢。也顺带堵掉一句最容易顺口说出来的话——没有谁「抓住了机会」，只是碰巧变异出硬一点外皮、大一点眼睛的那些个体，在一片有了捕食者的海里留下了更多后代。
EN The explanation you hear most often is oxygen: a big body that swims and chases things burns far more of it than one lying still, so this had to wait for the water to hold enough. Plausible, but the evidence does not point one way. Some geochemical proxies support a rise around then; other work finds oxygen going up and down in pulses rather than climbing one step; and a 2019 study that laid fossil and environmental records side by side argued the Ediacaran and Cambrian worlds were not as sharply distinct as the standard story has it. The safer reading is that oxygen was a background condition raising a ceiling, not a trigger you press. It is still argued over. The second explanation is more interesting: eyes. Trilobite eyes are among the earliest high-resolution eyes known, with lenses made of calcite crystal that can still be counted one by one on the fossil. Once something in the sea could see clearly a foot away, everything changed at once — hiding, playing dead, burying yourself, shells, spines, speed all became useful the same day, and better defence pushed predators to escalate in turn. The arms race from the piece on the Red Queen starts running here at scale. Could it have been new genetic tools arriving? That one is mostly ruled out: comparing animal genomes shows the regulatory genes that segment a body and decide what each segment becomes were already there before the Cambrian. The toolkit was old; what changed was the wiring and the opportunity outside. Lined up, these do not sort into first and second: seeing leads to catching, catching leads to shells and spines, which force better weapons, which make eyes worth more investment. Asking what "the cause" was is slightly the wrong question — it wants a trigger, and this looks more like a loop that started turning. Nobody seized an opportunity, either: individuals that happened to vary toward a harder skin or a bigger eye left more descendants in a sea that now contained predators.
寒武纪之后的四亿多年里，动物没有再添过新的基本样式。这件事比爆发本身更需要解释：变异一天也没停过，为什么样式的花名册就锁死了？
第一条解释是位置被占满了。在一个还没什么占位者的海里，一套只是凑合能用的新搭法也站得住脚——反正对手也粗糙。而今天任何一套新搭法，一出场就得跟已经被磨了五亿年的对手同场。门槛从来不是绝对水平，是相对于已经在场的那些。
第二条是改不动了。要换一套基本搭法，就得改早期发育。但这里有个常被说反的细节：不是越早的步骤越改不动。拿基因表达数据测下来，最保守的是中间那个阶段——各类动物的胚胎恰恰在那一刻长得最像；比它更早和更晚的阶段，反倒各走各的。这个形状被叫作沙漏，腰在中段。道理不难想：腰上那批基因被后面最多的过程同时依赖着，动一下，后面全塌。
第三条最清醒：「基本样式」这个级别是我们回头划的。那些支系刚分开时差别未必有多惊人，是后来各自又走了五亿年才积累成今天这个量级；早分开的支系必然显得更「根本」，这不需要任何特殊机制，是分类方式本身带来的。三条解释互不排斥，权重也没定论。
还有个著名说法后来被修正了。Gould 曾主张寒武纪那批生物的形态差异比今天还大，后来大批被抽签式地淘汰掉。现在多数人认为这话说过了头：不少当年的「怪物」被安回了已知类群。怪诞虫是最好的例子——它最初被复原成一只踩着七对硬刺走路、背上顶着一排触手的怪东西，1991 年才发现整个上下颠倒了（刺是背上的防御，触手才是腿），而且前后也反了，当初认作头的那端是尾。2014 年有人从它爪子里一层套一层的锥形结构，把它接到了天鹅绒虫那一支上。它不是一次失败的实验，它是有爪动物的亲戚。
EN In the four hundred million years since, animals have added no new basic body plan. That needs more explaining than the explosion does: variation never stopped, so why did the roster close? First answer: the seats filled up. In a sea with few incumbents, a merely workable new layout could hold its ground, because the competition was rough too. Any new layout today has to face opponents polished for half a billion years. The threshold was never an absolute standard; it is relative to whoever is already there. Second: it stopped being changeable. Swapping a body plan means altering early development — but with a detail people usually get backwards. It is not that the earliest steps are the hardest to change. Measured with gene-expression data, the most conserved stage is the middle one, where embryos of different animals look most alike; stages before and after are freer to diverge. The shape is called an hourglass, and the waist is in the middle, because the genes at the waist are the ones the largest number of later processes depend on at once. Third, and most sobering: the level we call a "basic body plan" is one we drew in hindsight. When those lineages first split, the differences need not have been dramatic; they became this large because each side then ran for another half a billion years. Early-branching lineages are bound to look more fundamental, and that follows from the way we classify, not from any special mechanism. The three are not exclusive and their weights are unsettled. One famous claim has since been trimmed: Gould argued that Cambrian animals spanned a wider range of forms than today's and that most were then culled by something close to a lottery. Most workers now think that went too far — many of the "weird wonders" have been placed back within known groups. Hallucigenia is the best example: first reconstructed walking on seven pairs of rigid spines with a row of tentacles on its back, it turned out in 1991 to be upside down (the spines are dorsal defence, the tentacles are the legs) and back to front, with the supposed head being the tail. In 2014 the nested cone-in-cone structure of its claws placed it on the branch leading to velvet worms. Not a failed experiment — a relative of the onychophorans.